3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.10046549010414922,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692002010345458,
|
|
"epoch": 0.00033488496701383075,
|
|
"grad_norm": 10.5625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.798,
|
|
"mean_token_accuracy": 0.00014925373252481223,
|
|
"num_tokens": 10113.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69187297821045,
|
|
"epoch": 0.0006697699340276615,
|
|
"grad_norm": 11.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.749,
|
|
"mean_token_accuracy": 0.000525744108017534,
|
|
"num_tokens": 19598.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690606594085693,
|
|
"epoch": 0.0010046549010414923,
|
|
"grad_norm": 8.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5809,
|
|
"mean_token_accuracy": 0.0301290619187057,
|
|
"num_tokens": 29705.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.670234203338623,
|
|
"epoch": 0.001339539868055323,
|
|
"grad_norm": 8.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3539,
|
|
"mean_token_accuracy": 0.04047959893941879,
|
|
"num_tokens": 39871.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.557568454742432,
|
|
"epoch": 0.0016744248350691537,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.1453,
|
|
"mean_token_accuracy": 0.04189771674573421,
|
|
"num_tokens": 49805.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.465806770324708,
|
|
"epoch": 0.0020093098020829846,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9507,
|
|
"mean_token_accuracy": 0.04296911805868149,
|
|
"num_tokens": 59845.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.461338233947753,
|
|
"epoch": 0.0023441947690968153,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9614,
|
|
"mean_token_accuracy": 0.0368592556566,
|
|
"num_tokens": 69836.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.483565521240234,
|
|
"epoch": 0.002679079736110646,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.8929,
|
|
"mean_token_accuracy": 0.046063635125756265,
|
|
"num_tokens": 80088.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.410326480865479,
|
|
"epoch": 0.0030139647031244767,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8768,
|
|
"mean_token_accuracy": 0.04229466691613197,
|
|
"num_tokens": 89458.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.434312725067139,
|
|
"epoch": 0.0033488496701383074,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.8494,
|
|
"mean_token_accuracy": 0.04813771955668926,
|
|
"num_tokens": 99523.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.504166984558106,
|
|
"epoch": 0.003683734637152138,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8242,
|
|
"mean_token_accuracy": 0.05405057594180107,
|
|
"num_tokens": 110080.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.344751262664795,
|
|
"epoch": 0.004018619604165969,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7129,
|
|
"mean_token_accuracy": 0.05208342745900154,
|
|
"num_tokens": 119412.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.211891746520996,
|
|
"epoch": 0.0043535045711798,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.6776,
|
|
"mean_token_accuracy": 0.051375290378928185,
|
|
"num_tokens": 129432.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.302197170257568,
|
|
"epoch": 0.0046883895381936306,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.6112,
|
|
"mean_token_accuracy": 0.059913215786218645,
|
|
"num_tokens": 139756.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.31943874359131,
|
|
"epoch": 0.005023274505207461,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5806,
|
|
"mean_token_accuracy": 0.05665694735944271,
|
|
"num_tokens": 148846.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.25365629196167,
|
|
"epoch": 0.005358159472221292,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.5087,
|
|
"mean_token_accuracy": 0.06129935123026371,
|
|
"num_tokens": 158647.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.23855152130127,
|
|
"epoch": 0.005693044439235123,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.3883,
|
|
"mean_token_accuracy": 0.060005851462483405,
|
|
"num_tokens": 167701.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.186964225769042,
|
|
"epoch": 0.006027929406248953,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3761,
|
|
"mean_token_accuracy": 0.06175476983189583,
|
|
"num_tokens": 178428.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.192708683013915,
|
|
"epoch": 0.006362814373262784,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3126,
|
|
"mean_token_accuracy": 0.057852984964847566,
|
|
"num_tokens": 187737.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.148875999450684,
|
|
"epoch": 0.006697699340276615,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3047,
|
|
"mean_token_accuracy": 0.05756702236831188,
|
|
"num_tokens": 198374.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.143085479736328,
|
|
"epoch": 0.007032584307290445,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2348,
|
|
"mean_token_accuracy": 0.05544475130736828,
|
|
"num_tokens": 208483.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.065252590179444,
|
|
"epoch": 0.007367469274304276,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1179,
|
|
"mean_token_accuracy": 0.0532416757196188,
|
|
"num_tokens": 218715.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.994503688812255,
|
|
"epoch": 0.007702354241318108,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9596,
|
|
"mean_token_accuracy": 0.06599014401435851,
|
|
"num_tokens": 227777.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.866777515411377,
|
|
"epoch": 0.008037239208331938,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.982,
|
|
"mean_token_accuracy": 0.05981399100273847,
|
|
"num_tokens": 238192.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.84700288772583,
|
|
"epoch": 0.008372124175345769,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.827,
|
|
"mean_token_accuracy": 0.06517648994922638,
|
|
"num_tokens": 247259.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.726642894744874,
|
|
"epoch": 0.0087070091423596,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8551,
|
|
"mean_token_accuracy": 0.054852675274014474,
|
|
"num_tokens": 258100.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.696339225769043,
|
|
"epoch": 0.00904189410937343,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.7497,
|
|
"mean_token_accuracy": 0.061625052243471146,
|
|
"num_tokens": 267620.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.46251974105835,
|
|
"epoch": 0.009376779076387261,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6451,
|
|
"mean_token_accuracy": 0.07104777507483959,
|
|
"num_tokens": 277654.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.472876358032227,
|
|
"epoch": 0.009711664043401092,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6622,
|
|
"mean_token_accuracy": 0.061386901512742044,
|
|
"num_tokens": 288195.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.343764114379884,
|
|
"epoch": 0.010046549010414923,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.6348,
|
|
"mean_token_accuracy": 0.0647747952491045,
|
|
"num_tokens": 298860.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.256326389312743,
|
|
"epoch": 0.010381433977428753,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.5569,
|
|
"mean_token_accuracy": 0.06752786003053188,
|
|
"num_tokens": 307799.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.192852115631103,
|
|
"epoch": 0.010716318944442584,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.5493,
|
|
"mean_token_accuracy": 0.06289746835827828,
|
|
"num_tokens": 318257.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.057830619812012,
|
|
"epoch": 0.011051203911456415,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.5351,
|
|
"mean_token_accuracy": 0.06436791345477104,
|
|
"num_tokens": 329065.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.072613048553468,
|
|
"epoch": 0.011386088878470245,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4825,
|
|
"mean_token_accuracy": 0.06475077606737614,
|
|
"num_tokens": 338809.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.959860801696777,
|
|
"epoch": 0.011720973845484076,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.4588,
|
|
"mean_token_accuracy": 0.06541862487792968,
|
|
"num_tokens": 349318.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.849260616302491,
|
|
"epoch": 0.012055858812497907,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.4348,
|
|
"mean_token_accuracy": 0.06766009479761123,
|
|
"num_tokens": 358760.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.860241985321045,
|
|
"epoch": 0.012390743779511737,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.3168,
|
|
"mean_token_accuracy": 0.06955580413341522,
|
|
"num_tokens": 367589.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.835146236419678,
|
|
"epoch": 0.012725628746525568,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.3624,
|
|
"mean_token_accuracy": 0.06997261717915534,
|
|
"num_tokens": 377012.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.665368556976318,
|
|
"epoch": 0.013060513713539399,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.3276,
|
|
"mean_token_accuracy": 0.07220411114394665,
|
|
"num_tokens": 386791.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.820245933532714,
|
|
"epoch": 0.01339539868055323,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.4404,
|
|
"mean_token_accuracy": 0.06381001807749272,
|
|
"num_tokens": 396620.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.68319730758667,
|
|
"epoch": 0.01373028364756706,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.3971,
|
|
"mean_token_accuracy": 0.06604428850114345,
|
|
"num_tokens": 406384.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.685719966888428,
|
|
"epoch": 0.01406516861458089,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.3622,
|
|
"mean_token_accuracy": 0.07016287520527839,
|
|
"num_tokens": 416481.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.711257934570312,
|
|
"epoch": 0.014400053581594721,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.3841,
|
|
"mean_token_accuracy": 0.07033630795776843,
|
|
"num_tokens": 425697.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.652897453308105,
|
|
"epoch": 0.014734938548608552,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.2995,
|
|
"mean_token_accuracy": 0.0704294566065073,
|
|
"num_tokens": 435836.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.68178768157959,
|
|
"epoch": 0.015069823515622383,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.3569,
|
|
"mean_token_accuracy": 0.06939754635095596,
|
|
"num_tokens": 445926.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.638516330718994,
|
|
"epoch": 0.015404708482636215,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.2983,
|
|
"mean_token_accuracy": 0.07235002666711807,
|
|
"num_tokens": 455594.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.570248413085938,
|
|
"epoch": 0.015739593449650046,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.3461,
|
|
"mean_token_accuracy": 0.0698947947472334,
|
|
"num_tokens": 465928.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.62375259399414,
|
|
"epoch": 0.016074478416663877,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.2851,
|
|
"mean_token_accuracy": 0.07294886186718941,
|
|
"num_tokens": 475946.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.62733097076416,
|
|
"epoch": 0.016409363383677707,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.2878,
|
|
"mean_token_accuracy": 0.07264770828187465,
|
|
"num_tokens": 485285.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.576191997528076,
|
|
"epoch": 0.016744248350691538,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.2942,
|
|
"mean_token_accuracy": 0.07046032212674617,
|
|
"num_tokens": 496516.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.584659576416016,
|
|
"epoch": 0.01707913331770537,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.2829,
|
|
"mean_token_accuracy": 0.07218111529946328,
|
|
"num_tokens": 506330.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.551284503936767,
|
|
"epoch": 0.0174140182847192,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.3443,
|
|
"mean_token_accuracy": 0.07309619709849358,
|
|
"num_tokens": 515825.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.580291271209717,
|
|
"epoch": 0.01774890325173303,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2687,
|
|
"mean_token_accuracy": 0.06971881128847598,
|
|
"num_tokens": 526303.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.48112440109253,
|
|
"epoch": 0.01808378821874686,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.3759,
|
|
"mean_token_accuracy": 0.07506775110960007,
|
|
"num_tokens": 536805.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.600781631469726,
|
|
"epoch": 0.01841867318576069,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.2216,
|
|
"mean_token_accuracy": 0.07659760937094688,
|
|
"num_tokens": 546381.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.565075874328613,
|
|
"epoch": 0.018753558152774522,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.2444,
|
|
"mean_token_accuracy": 0.0764175396412611,
|
|
"num_tokens": 556278.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.52633352279663,
|
|
"epoch": 0.019088443119788353,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2569,
|
|
"mean_token_accuracy": 0.07796998918056489,
|
|
"num_tokens": 565312.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.471183109283448,
|
|
"epoch": 0.019423328086802184,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.2313,
|
|
"mean_token_accuracy": 0.07882754877209663,
|
|
"num_tokens": 575304.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.441308689117431,
|
|
"epoch": 0.019758213053816014,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.1941,
|
|
"mean_token_accuracy": 0.07772204428911209,
|
|
"num_tokens": 585262.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.575240802764892,
|
|
"epoch": 0.020093098020829845,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.2345,
|
|
"mean_token_accuracy": 0.07687565684318542,
|
|
"num_tokens": 595297.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.531679725646972,
|
|
"epoch": 0.020427982987843676,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.2113,
|
|
"mean_token_accuracy": 0.07619660049676895,
|
|
"num_tokens": 604354.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.408655452728272,
|
|
"epoch": 0.020762867954857506,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.3286,
|
|
"mean_token_accuracy": 0.07043087370693683,
|
|
"num_tokens": 616283.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.580455112457276,
|
|
"epoch": 0.021097752921871337,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.2259,
|
|
"mean_token_accuracy": 0.07452421076595783,
|
|
"num_tokens": 626059.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.43509645462036,
|
|
"epoch": 0.021432637888885168,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.1612,
|
|
"mean_token_accuracy": 0.07599973231554032,
|
|
"num_tokens": 634483.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.409455680847168,
|
|
"epoch": 0.021767522855899,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.1607,
|
|
"mean_token_accuracy": 0.08002224043011666,
|
|
"num_tokens": 643111.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.471851444244384,
|
|
"epoch": 0.02210240782291283,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.1549,
|
|
"mean_token_accuracy": 0.08036141917109489,
|
|
"num_tokens": 653650.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.432329463958741,
|
|
"epoch": 0.02243729278992666,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1509,
|
|
"mean_token_accuracy": 0.08708992898464203,
|
|
"num_tokens": 663143.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.438332271575927,
|
|
"epoch": 0.02277217775694049,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.1221,
|
|
"mean_token_accuracy": 0.08310596123337746,
|
|
"num_tokens": 672434.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.37921953201294,
|
|
"epoch": 0.02310706272395432,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.1224,
|
|
"mean_token_accuracy": 0.07925173193216324,
|
|
"num_tokens": 682973.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.429475784301758,
|
|
"epoch": 0.023441947690968152,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.1876,
|
|
"mean_token_accuracy": 0.07883969098329544,
|
|
"num_tokens": 692136.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.41495771408081,
|
|
"epoch": 0.023776832657981983,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.0919,
|
|
"mean_token_accuracy": 0.08360464051365853,
|
|
"num_tokens": 702257.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.438455581665039,
|
|
"epoch": 0.024111717624995813,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1034,
|
|
"mean_token_accuracy": 0.08223021924495696,
|
|
"num_tokens": 712444.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.402928638458253,
|
|
"epoch": 0.024446602592009644,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.2461,
|
|
"mean_token_accuracy": 0.07820370681583881,
|
|
"num_tokens": 721614.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.494019985198975,
|
|
"epoch": 0.024781487559023475,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.2242,
|
|
"mean_token_accuracy": 0.07433447390794753,
|
|
"num_tokens": 731189.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.423804473876952,
|
|
"epoch": 0.025116372526037305,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.1387,
|
|
"mean_token_accuracy": 0.08087190091609955,
|
|
"num_tokens": 741555.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.242571067810058,
|
|
"epoch": 0.025451257493051136,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.0677,
|
|
"mean_token_accuracy": 0.084798514097929,
|
|
"num_tokens": 751560.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.45106019973755,
|
|
"epoch": 0.025786142460064967,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.1193,
|
|
"mean_token_accuracy": 0.07877458333969116,
|
|
"num_tokens": 761471.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.435086345672607,
|
|
"epoch": 0.026121027427078797,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.1184,
|
|
"mean_token_accuracy": 0.08121965527534485,
|
|
"num_tokens": 771565.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.258657455444336,
|
|
"epoch": 0.026455912394092628,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.0685,
|
|
"mean_token_accuracy": 0.08140704035758972,
|
|
"num_tokens": 781248.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.397636890411377,
|
|
"epoch": 0.02679079736110646,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.1233,
|
|
"mean_token_accuracy": 0.0783721823245287,
|
|
"num_tokens": 790907.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.232399845123291,
|
|
"epoch": 0.02712568232812029,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.1384,
|
|
"mean_token_accuracy": 0.07922295071184635,
|
|
"num_tokens": 800073.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.517311382293702,
|
|
"epoch": 0.02746056729513412,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.099,
|
|
"mean_token_accuracy": 0.0789324052631855,
|
|
"num_tokens": 809782.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.192907619476319,
|
|
"epoch": 0.02779545226214795,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0579,
|
|
"mean_token_accuracy": 0.08063366934657097,
|
|
"num_tokens": 820329.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.31554536819458,
|
|
"epoch": 0.02813033722916178,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9757,
|
|
"mean_token_accuracy": 0.09008945301175117,
|
|
"num_tokens": 829405.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.286150169372558,
|
|
"epoch": 0.028465222196175612,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.0418,
|
|
"mean_token_accuracy": 0.08079335764050484,
|
|
"num_tokens": 838700.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.263776969909667,
|
|
"epoch": 0.028800107163189443,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.1043,
|
|
"mean_token_accuracy": 0.083824248239398,
|
|
"num_tokens": 848892.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.295380115509033,
|
|
"epoch": 0.029134992130203274,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0354,
|
|
"mean_token_accuracy": 0.08374309539794922,
|
|
"num_tokens": 857737.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.229405975341797,
|
|
"epoch": 0.029469877097217104,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.968,
|
|
"mean_token_accuracy": 0.08430874347686768,
|
|
"num_tokens": 867758.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.303557300567627,
|
|
"epoch": 0.029804762064230935,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.9853,
|
|
"mean_token_accuracy": 0.08112291395664215,
|
|
"num_tokens": 877273.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.286294078826904,
|
|
"epoch": 0.030139647031244766,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.087,
|
|
"mean_token_accuracy": 0.08337052017450333,
|
|
"num_tokens": 887109.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.264797973632813,
|
|
"epoch": 0.030474531998258596,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.0057,
|
|
"mean_token_accuracy": 0.08394731357693672,
|
|
"num_tokens": 896157.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.19506025314331,
|
|
"epoch": 0.03080941696527243,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.011,
|
|
"mean_token_accuracy": 0.07882361188530922,
|
|
"num_tokens": 907231.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.185263442993165,
|
|
"epoch": 0.03114430193228626,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.9223,
|
|
"mean_token_accuracy": 0.0873487912118435,
|
|
"num_tokens": 916756.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.243685722351074,
|
|
"epoch": 0.03147918689930009,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.0922,
|
|
"mean_token_accuracy": 0.07646922953426838,
|
|
"num_tokens": 927714.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.29259443283081,
|
|
"epoch": 0.03181407186631392,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.9389,
|
|
"mean_token_accuracy": 0.08775902166962624,
|
|
"num_tokens": 936620.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.152761650085449,
|
|
"epoch": 0.03214895683332775,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.9753,
|
|
"mean_token_accuracy": 0.08129271119832993,
|
|
"num_tokens": 945942.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.177737617492676,
|
|
"epoch": 0.032483841800341584,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.0066,
|
|
"mean_token_accuracy": 0.08826668187975883,
|
|
"num_tokens": 956685.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.196662044525146,
|
|
"epoch": 0.032818726767355415,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.9442,
|
|
"mean_token_accuracy": 0.08767017722129822,
|
|
"num_tokens": 967453.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.19489345550537,
|
|
"epoch": 0.033153611734369245,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9764,
|
|
"mean_token_accuracy": 0.08438041880726814,
|
|
"num_tokens": 978220.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.135752773284912,
|
|
"epoch": 0.033488496701383076,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9185,
|
|
"mean_token_accuracy": 0.09252278879284859,
|
|
"num_tokens": 988120.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.218993663787842,
|
|
"epoch": 0.03382338166839691,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.9388,
|
|
"mean_token_accuracy": 0.08044274970889091,
|
|
"num_tokens": 997733.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.14416561126709,
|
|
"epoch": 0.03415826663541074,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9004,
|
|
"mean_token_accuracy": 0.08713611736893653,
|
|
"num_tokens": 1007362.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.12597336769104,
|
|
"epoch": 0.03449315160242457,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.8522,
|
|
"mean_token_accuracy": 0.08573579043149948,
|
|
"num_tokens": 1016352.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.978172826766968,
|
|
"epoch": 0.0348280365694384,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.8683,
|
|
"mean_token_accuracy": 0.08755372166633606,
|
|
"num_tokens": 1025944.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.23288698196411,
|
|
"epoch": 0.03516292153645223,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.9434,
|
|
"mean_token_accuracy": 0.08828524276614189,
|
|
"num_tokens": 1035675.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.088372182846069,
|
|
"epoch": 0.03549780650346606,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.936,
|
|
"mean_token_accuracy": 0.08637615889310837,
|
|
"num_tokens": 1045944.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.136125946044922,
|
|
"epoch": 0.03583269147047989,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.8464,
|
|
"mean_token_accuracy": 0.09038148149847984,
|
|
"num_tokens": 1055721.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.160592937469483,
|
|
"epoch": 0.03616757643749372,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.978,
|
|
"mean_token_accuracy": 0.08611000180244446,
|
|
"num_tokens": 1065841.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.033481216430664,
|
|
"epoch": 0.03650246140450755,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.976,
|
|
"mean_token_accuracy": 0.08632155880331993,
|
|
"num_tokens": 1076574.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.213283681869507,
|
|
"epoch": 0.03683734637152138,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.9227,
|
|
"mean_token_accuracy": 0.08273187503218651,
|
|
"num_tokens": 1085761.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.176979160308838,
|
|
"epoch": 0.037172231338535214,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.9048,
|
|
"mean_token_accuracy": 0.08671903014183044,
|
|
"num_tokens": 1096036.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.981817817687988,
|
|
"epoch": 0.037507116305549044,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8203,
|
|
"mean_token_accuracy": 0.08757113367319107,
|
|
"num_tokens": 1105103.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.080931758880615,
|
|
"epoch": 0.037842001272562875,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.863,
|
|
"mean_token_accuracy": 0.09306468516588211,
|
|
"num_tokens": 1115355.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.037179470062256,
|
|
"epoch": 0.038176886239576706,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8377,
|
|
"mean_token_accuracy": 0.08505469113588333,
|
|
"num_tokens": 1125977.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.067077589035033,
|
|
"epoch": 0.03851177120659054,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.8574,
|
|
"mean_token_accuracy": 0.08890697509050369,
|
|
"num_tokens": 1135536.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.069751501083374,
|
|
"epoch": 0.03884665617360437,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.7972,
|
|
"mean_token_accuracy": 0.0888807274401188,
|
|
"num_tokens": 1144857.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.992953777313232,
|
|
"epoch": 0.0391815411406182,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.8085,
|
|
"mean_token_accuracy": 0.08944643214344979,
|
|
"num_tokens": 1154188.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.046015882492066,
|
|
"epoch": 0.03951642610763203,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.847,
|
|
"mean_token_accuracy": 0.08940242640674115,
|
|
"num_tokens": 1164583.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.920463609695434,
|
|
"epoch": 0.03985131107464586,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.8122,
|
|
"mean_token_accuracy": 0.09227430075407028,
|
|
"num_tokens": 1174163.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.987206220626831,
|
|
"epoch": 0.04018619604165969,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.8286,
|
|
"mean_token_accuracy": 0.08923885226249695,
|
|
"num_tokens": 1184065.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.923127269744873,
|
|
"epoch": 0.04052108100867352,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.6988,
|
|
"mean_token_accuracy": 0.1023092582821846,
|
|
"num_tokens": 1193263.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.989051961898804,
|
|
"epoch": 0.04085596597568735,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.7558,
|
|
"mean_token_accuracy": 0.09200508967041969,
|
|
"num_tokens": 1202940.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.819897556304932,
|
|
"epoch": 0.04119085094270118,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.6977,
|
|
"mean_token_accuracy": 0.09320536255836487,
|
|
"num_tokens": 1212978.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.914996862411499,
|
|
"epoch": 0.04152573590971501,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7375,
|
|
"mean_token_accuracy": 0.09381118640303612,
|
|
"num_tokens": 1222945.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.922825765609741,
|
|
"epoch": 0.04186062087672884,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.7243,
|
|
"mean_token_accuracy": 0.0925237312912941,
|
|
"num_tokens": 1232041.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.018640756607056,
|
|
"epoch": 0.042195505843742674,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.7608,
|
|
"mean_token_accuracy": 0.09631229490041733,
|
|
"num_tokens": 1241765.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.8229437351226805,
|
|
"epoch": 0.042530390810756505,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.7353,
|
|
"mean_token_accuracy": 0.09919300377368927,
|
|
"num_tokens": 1250642.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.911471843719482,
|
|
"epoch": 0.042865275777770336,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.7431,
|
|
"mean_token_accuracy": 0.09453340247273445,
|
|
"num_tokens": 1260250.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.082605934143066,
|
|
"epoch": 0.043200160744784166,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.8231,
|
|
"mean_token_accuracy": 0.08769241273403168,
|
|
"num_tokens": 1269786.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.858733367919922,
|
|
"epoch": 0.043535045711798,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7604,
|
|
"mean_token_accuracy": 0.08618547506630421,
|
|
"num_tokens": 1280029.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.007468605041504,
|
|
"epoch": 0.04386993067881183,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.718,
|
|
"mean_token_accuracy": 0.08944717943668365,
|
|
"num_tokens": 1288983.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.869023227691651,
|
|
"epoch": 0.04420481564582566,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7681,
|
|
"mean_token_accuracy": 0.09466379135847092,
|
|
"num_tokens": 1298629.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.920618724822998,
|
|
"epoch": 0.04453970061283949,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.7448,
|
|
"mean_token_accuracy": 0.08919373750686646,
|
|
"num_tokens": 1308611.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.893028736114502,
|
|
"epoch": 0.04487458557985332,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7868,
|
|
"mean_token_accuracy": 0.08869361504912376,
|
|
"num_tokens": 1318150.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.96576623916626,
|
|
"epoch": 0.04520947054686715,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7329,
|
|
"mean_token_accuracy": 0.0911085106432438,
|
|
"num_tokens": 1328933.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.807632970809936,
|
|
"epoch": 0.04554435551388098,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7145,
|
|
"mean_token_accuracy": 0.09275061488151551,
|
|
"num_tokens": 1338614.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.796882915496826,
|
|
"epoch": 0.04587924048089481,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6533,
|
|
"mean_token_accuracy": 0.08674826323986054,
|
|
"num_tokens": 1348292.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.877967834472656,
|
|
"epoch": 0.04621412544790864,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7007,
|
|
"mean_token_accuracy": 0.09283162206411362,
|
|
"num_tokens": 1358479.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.635259914398193,
|
|
"epoch": 0.04654901041492247,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.5801,
|
|
"mean_token_accuracy": 0.09727387279272079,
|
|
"num_tokens": 1367778.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.858510446548462,
|
|
"epoch": 0.046883895381936304,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.668,
|
|
"mean_token_accuracy": 0.09592612758278847,
|
|
"num_tokens": 1376826.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.8318394184112545,
|
|
"epoch": 0.047218780348950135,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7229,
|
|
"mean_token_accuracy": 0.09366490095853805,
|
|
"num_tokens": 1386280.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.815138149261474,
|
|
"epoch": 0.047553665315963965,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7523,
|
|
"mean_token_accuracy": 0.09019050747156143,
|
|
"num_tokens": 1396062.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.9230225563049315,
|
|
"epoch": 0.047888550282977796,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.7367,
|
|
"mean_token_accuracy": 0.0929199367761612,
|
|
"num_tokens": 1406980.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.8387205600738525,
|
|
"epoch": 0.04822343524999163,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.7349,
|
|
"mean_token_accuracy": 0.09628795683383942,
|
|
"num_tokens": 1417341.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.781990957260132,
|
|
"epoch": 0.04855832021700546,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.7154,
|
|
"mean_token_accuracy": 0.09174965173006058,
|
|
"num_tokens": 1427066.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.9304931640625,
|
|
"epoch": 0.04889320518401929,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6933,
|
|
"mean_token_accuracy": 0.0906587302684784,
|
|
"num_tokens": 1437429.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.806986618041992,
|
|
"epoch": 0.04922809015103312,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.795,
|
|
"mean_token_accuracy": 0.08047471344470977,
|
|
"num_tokens": 1449128.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.767903661727905,
|
|
"epoch": 0.04956297511804695,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6489,
|
|
"mean_token_accuracy": 0.09558213129639626,
|
|
"num_tokens": 1459397.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.746322727203369,
|
|
"epoch": 0.04989786008506078,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6609,
|
|
"mean_token_accuracy": 0.09293203428387642,
|
|
"num_tokens": 1469424.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.720838212966919,
|
|
"epoch": 0.05023274505207461,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6218,
|
|
"mean_token_accuracy": 0.10069447308778763,
|
|
"num_tokens": 1478980.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.836196994781494,
|
|
"epoch": 0.05056763001908844,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.6726,
|
|
"mean_token_accuracy": 0.09147711992263793,
|
|
"num_tokens": 1488655.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.671448421478272,
|
|
"epoch": 0.05090251498610227,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.6464,
|
|
"mean_token_accuracy": 0.0919630341231823,
|
|
"num_tokens": 1498248.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.777037143707275,
|
|
"epoch": 0.0512373999531161,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6067,
|
|
"mean_token_accuracy": 0.0970597043633461,
|
|
"num_tokens": 1509496.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.764213752746582,
|
|
"epoch": 0.051572284920129934,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.6505,
|
|
"mean_token_accuracy": 0.0956808902323246,
|
|
"num_tokens": 1518741.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.794732761383057,
|
|
"epoch": 0.051907169887143764,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.6665,
|
|
"mean_token_accuracy": 0.09552088975906373,
|
|
"num_tokens": 1528928.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.712340307235718,
|
|
"epoch": 0.052242054854157595,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5846,
|
|
"mean_token_accuracy": 0.09882296845316887,
|
|
"num_tokens": 1538255.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.661662578582764,
|
|
"epoch": 0.052576939821171426,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.7101,
|
|
"mean_token_accuracy": 0.09138747528195382,
|
|
"num_tokens": 1549014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.798562860488891,
|
|
"epoch": 0.052911824788185256,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.5892,
|
|
"mean_token_accuracy": 0.09761035665869713,
|
|
"num_tokens": 1558905.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.620625352859497,
|
|
"epoch": 0.05324670975519909,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.5663,
|
|
"mean_token_accuracy": 0.09829202219843865,
|
|
"num_tokens": 1568605.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.721243143081665,
|
|
"epoch": 0.05358159472221292,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.5793,
|
|
"mean_token_accuracy": 0.09484488666057586,
|
|
"num_tokens": 1579330.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.629036903381348,
|
|
"epoch": 0.05391647968922675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5517,
|
|
"mean_token_accuracy": 0.10525800734758377,
|
|
"num_tokens": 1588702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.6774791240692135,
|
|
"epoch": 0.05425136465624058,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5431,
|
|
"mean_token_accuracy": 0.09909340515732765,
|
|
"num_tokens": 1598789.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.629120206832885,
|
|
"epoch": 0.05458624962325441,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.5682,
|
|
"mean_token_accuracy": 0.09840805605053901,
|
|
"num_tokens": 1608580.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.713317251205444,
|
|
"epoch": 0.05492113459026824,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5903,
|
|
"mean_token_accuracy": 0.09436297565698623,
|
|
"num_tokens": 1618344.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.623961067199707,
|
|
"epoch": 0.05525601955728207,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.5522,
|
|
"mean_token_accuracy": 0.09576157703995705,
|
|
"num_tokens": 1627654.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.673608446121216,
|
|
"epoch": 0.0555909045242959,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.7425,
|
|
"mean_token_accuracy": 0.09187466725707054,
|
|
"num_tokens": 1638591.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.863967800140381,
|
|
"epoch": 0.05592578949130973,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.7303,
|
|
"mean_token_accuracy": 0.08798938393592834,
|
|
"num_tokens": 1649384.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.7440564155578615,
|
|
"epoch": 0.05626067445832356,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.7142,
|
|
"mean_token_accuracy": 0.08833635300397873,
|
|
"num_tokens": 1659079.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.69318208694458,
|
|
"epoch": 0.056595559425337394,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.6531,
|
|
"mean_token_accuracy": 0.09180667996406555,
|
|
"num_tokens": 1669245.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.693623304367065,
|
|
"epoch": 0.056930444392351225,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.5619,
|
|
"mean_token_accuracy": 0.09287880584597588,
|
|
"num_tokens": 1678781.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.553192472457885,
|
|
"epoch": 0.057265329359365055,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5305,
|
|
"mean_token_accuracy": 0.09830001443624496,
|
|
"num_tokens": 1688192.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.70746922492981,
|
|
"epoch": 0.057600214326378886,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5205,
|
|
"mean_token_accuracy": 0.09981869980692863,
|
|
"num_tokens": 1698058.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.522954750061035,
|
|
"epoch": 0.05793509929339272,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5455,
|
|
"mean_token_accuracy": 0.09677255153656006,
|
|
"num_tokens": 1708568.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.559320640563965,
|
|
"epoch": 0.05826998426040655,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.4243,
|
|
"mean_token_accuracy": 0.10318205356597901,
|
|
"num_tokens": 1718854.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.73996000289917,
|
|
"epoch": 0.05860486922742038,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.5525,
|
|
"mean_token_accuracy": 0.09444483071565628,
|
|
"num_tokens": 1728549.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.554740953445434,
|
|
"epoch": 0.05893975419443421,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5246,
|
|
"mean_token_accuracy": 0.09602352306246757,
|
|
"num_tokens": 1737664.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.663396739959717,
|
|
"epoch": 0.05927463916144804,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.6325,
|
|
"mean_token_accuracy": 0.09179740771651268,
|
|
"num_tokens": 1748791.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.649189853668213,
|
|
"epoch": 0.05960952412846187,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5579,
|
|
"mean_token_accuracy": 0.09603997468948364,
|
|
"num_tokens": 1758323.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.5717028141021725,
|
|
"epoch": 0.0599444090954757,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5289,
|
|
"mean_token_accuracy": 0.09905452281236649,
|
|
"num_tokens": 1768992.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.641081666946411,
|
|
"epoch": 0.06027929406248953,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.5959,
|
|
"mean_token_accuracy": 0.09036346450448036,
|
|
"num_tokens": 1779810.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.533492517471314,
|
|
"epoch": 0.06061417902950336,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4949,
|
|
"mean_token_accuracy": 0.10070249736309052,
|
|
"num_tokens": 1790071.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.578271436691284,
|
|
"epoch": 0.06094906399651719,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.5443,
|
|
"mean_token_accuracy": 0.09873676598072052,
|
|
"num_tokens": 1799818.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.532402181625367,
|
|
"epoch": 0.06128394896353103,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.5053,
|
|
"mean_token_accuracy": 0.09719008877873421,
|
|
"num_tokens": 1810559.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.512709522247315,
|
|
"epoch": 0.06161883393054486,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4387,
|
|
"mean_token_accuracy": 0.09774044081568718,
|
|
"num_tokens": 1819810.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.604777812957764,
|
|
"epoch": 0.06195371889755869,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.5699,
|
|
"mean_token_accuracy": 0.0973470464348793,
|
|
"num_tokens": 1830737.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.58216872215271,
|
|
"epoch": 0.06228860386457252,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.5539,
|
|
"mean_token_accuracy": 0.09564543291926383,
|
|
"num_tokens": 1841134.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.623823022842407,
|
|
"epoch": 0.06262348883158635,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4891,
|
|
"mean_token_accuracy": 0.1037807047367096,
|
|
"num_tokens": 1851962.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.517935132980346,
|
|
"epoch": 0.06295837379860018,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.5686,
|
|
"mean_token_accuracy": 0.09641919285058975,
|
|
"num_tokens": 1862748.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.5019701480865475,
|
|
"epoch": 0.06329325876561401,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.5053,
|
|
"mean_token_accuracy": 0.09677049964666366,
|
|
"num_tokens": 1872398.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.4540633201599125,
|
|
"epoch": 0.06362814373262785,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.3761,
|
|
"mean_token_accuracy": 0.10685181841254235,
|
|
"num_tokens": 1882263.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.627192592620849,
|
|
"epoch": 0.06396302869964167,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.5367,
|
|
"mean_token_accuracy": 0.09033481255173684,
|
|
"num_tokens": 1893007.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.4436921119689945,
|
|
"epoch": 0.0642979136666555,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.5383,
|
|
"mean_token_accuracy": 0.10231668278574943,
|
|
"num_tokens": 1903357.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.596249151229858,
|
|
"epoch": 0.06463279863366933,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.4762,
|
|
"mean_token_accuracy": 0.09526494145393372,
|
|
"num_tokens": 1913793.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.469473552703858,
|
|
"epoch": 0.06496768360068317,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.452,
|
|
"mean_token_accuracy": 0.09744430407881736,
|
|
"num_tokens": 1922540.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.527045965194702,
|
|
"epoch": 0.06530256856769699,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.5215,
|
|
"mean_token_accuracy": 0.09597379714250565,
|
|
"num_tokens": 1933706.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.437564373016357,
|
|
"epoch": 0.06563745353471083,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.4096,
|
|
"mean_token_accuracy": 0.10182152763009071,
|
|
"num_tokens": 1944169.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.391198968887329,
|
|
"epoch": 0.06597233850172465,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.3428,
|
|
"mean_token_accuracy": 0.1063389353454113,
|
|
"num_tokens": 1954017.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.518598890304565,
|
|
"epoch": 0.06630722346873849,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5243,
|
|
"mean_token_accuracy": 0.0978463925421238,
|
|
"num_tokens": 1965075.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.42684416770935,
|
|
"epoch": 0.06664210843575231,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.4063,
|
|
"mean_token_accuracy": 0.10326866209506988,
|
|
"num_tokens": 1974808.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.512786722183227,
|
|
"epoch": 0.06697699340276615,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.4755,
|
|
"mean_token_accuracy": 0.09493202716112137,
|
|
"num_tokens": 1985172.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.5474714756011965,
|
|
"epoch": 0.06731187836977998,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.4858,
|
|
"mean_token_accuracy": 0.10103376507759095,
|
|
"num_tokens": 1996338.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.445226097106934,
|
|
"epoch": 0.06764676333679381,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.4732,
|
|
"mean_token_accuracy": 0.09459194168448448,
|
|
"num_tokens": 2005429.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.482739734649658,
|
|
"epoch": 0.06798164830380764,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.4646,
|
|
"mean_token_accuracy": 0.09422318562865258,
|
|
"num_tokens": 2015278.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.479223680496216,
|
|
"epoch": 0.06831653327082148,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.4532,
|
|
"mean_token_accuracy": 0.09788647592067719,
|
|
"num_tokens": 2025758.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.373387289047241,
|
|
"epoch": 0.0686514182378353,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.3723,
|
|
"mean_token_accuracy": 0.10488328263163567,
|
|
"num_tokens": 2035029.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.414871549606323,
|
|
"epoch": 0.06898630320484914,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.3718,
|
|
"mean_token_accuracy": 0.10269584804773331,
|
|
"num_tokens": 2044182.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.479082298278809,
|
|
"epoch": 0.06932118817186296,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.4598,
|
|
"mean_token_accuracy": 0.09844749867916107,
|
|
"num_tokens": 2054676.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.414534330368042,
|
|
"epoch": 0.0696560731388768,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.3981,
|
|
"mean_token_accuracy": 0.09748629033565522,
|
|
"num_tokens": 2062933.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.395028114318848,
|
|
"epoch": 0.06999095810589062,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.3612,
|
|
"mean_token_accuracy": 0.10393474400043487,
|
|
"num_tokens": 2073146.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.429117488861084,
|
|
"epoch": 0.07032584307290446,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.4926,
|
|
"mean_token_accuracy": 0.09564846009016037,
|
|
"num_tokens": 2083847.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.4130007266998295,
|
|
"epoch": 0.07066072803991828,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.3997,
|
|
"mean_token_accuracy": 0.09807902723550796,
|
|
"num_tokens": 2094969.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.458782052993774,
|
|
"epoch": 0.07099561300693212,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.4067,
|
|
"mean_token_accuracy": 0.0952724613249302,
|
|
"num_tokens": 2105275.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.375722408294678,
|
|
"epoch": 0.07133049797394594,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.4499,
|
|
"mean_token_accuracy": 0.09948632344603539,
|
|
"num_tokens": 2115833.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.466668558120728,
|
|
"epoch": 0.07166538294095978,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 7.3751,
|
|
"mean_token_accuracy": 0.1037701740860939,
|
|
"num_tokens": 2124747.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.423423814773559,
|
|
"epoch": 0.0720002679079736,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 7.4149,
|
|
"mean_token_accuracy": 0.09693367630243302,
|
|
"num_tokens": 2135114.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.397886610031128,
|
|
"epoch": 0.07233515287498744,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 7.433,
|
|
"mean_token_accuracy": 0.09895812720060349,
|
|
"num_tokens": 2144324.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.453474044799805,
|
|
"epoch": 0.07267003784200127,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 7.3563,
|
|
"mean_token_accuracy": 0.09745102897286415,
|
|
"num_tokens": 2153971.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.311913967132568,
|
|
"epoch": 0.0730049228090151,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 7.3283,
|
|
"mean_token_accuracy": 0.10133487731218338,
|
|
"num_tokens": 2164389.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.367936801910401,
|
|
"epoch": 0.07333980777602893,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 7.3145,
|
|
"mean_token_accuracy": 0.10260041207075118,
|
|
"num_tokens": 2173802.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.433255243301391,
|
|
"epoch": 0.07367469274304277,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 7.3249,
|
|
"mean_token_accuracy": 0.103522939234972,
|
|
"num_tokens": 2182590.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.250359439849854,
|
|
"epoch": 0.07400957771005659,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 7.3198,
|
|
"mean_token_accuracy": 0.1075370505452156,
|
|
"num_tokens": 2192025.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.5280996799469,
|
|
"epoch": 0.07434446267707043,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 7.4923,
|
|
"mean_token_accuracy": 0.10366464406251907,
|
|
"num_tokens": 2201999.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.378125333786011,
|
|
"epoch": 0.07467934764408425,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 7.3691,
|
|
"mean_token_accuracy": 0.09555793032050133,
|
|
"num_tokens": 2211804.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.355848169326782,
|
|
"epoch": 0.07501423261109809,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 7.2508,
|
|
"mean_token_accuracy": 0.1032866396009922,
|
|
"num_tokens": 2222079.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.278160572052002,
|
|
"epoch": 0.07534911757811191,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 7.3811,
|
|
"mean_token_accuracy": 0.10743188410997391,
|
|
"num_tokens": 2232213.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.39425253868103,
|
|
"epoch": 0.07568400254512575,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 7.3066,
|
|
"mean_token_accuracy": 0.10472791343927383,
|
|
"num_tokens": 2242017.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.2768415927886965,
|
|
"epoch": 0.07601888751213957,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 7.2968,
|
|
"mean_token_accuracy": 0.10491021052002907,
|
|
"num_tokens": 2251654.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.349087905883789,
|
|
"epoch": 0.07635377247915341,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 7.3466,
|
|
"mean_token_accuracy": 0.10458370298147202,
|
|
"num_tokens": 2262972.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.255107593536377,
|
|
"epoch": 0.07668865744616724,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 7.3488,
|
|
"mean_token_accuracy": 0.09580066353082657,
|
|
"num_tokens": 2273067.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.342799282073974,
|
|
"epoch": 0.07702354241318107,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 7.2318,
|
|
"mean_token_accuracy": 0.10805003568530083,
|
|
"num_tokens": 2282104.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.299344301223755,
|
|
"epoch": 0.0773584273801949,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 7.3451,
|
|
"mean_token_accuracy": 0.10275197774171829,
|
|
"num_tokens": 2291366.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.345579051971436,
|
|
"epoch": 0.07769331234720873,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 7.2663,
|
|
"mean_token_accuracy": 0.10576335266232491,
|
|
"num_tokens": 2300107.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.265869903564453,
|
|
"epoch": 0.07802819731422256,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 7.2317,
|
|
"mean_token_accuracy": 0.1063105285167694,
|
|
"num_tokens": 2309755.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.330099105834961,
|
|
"epoch": 0.0783630822812364,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 7.3643,
|
|
"mean_token_accuracy": 0.09808892235159875,
|
|
"num_tokens": 2320042.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.272184801101685,
|
|
"epoch": 0.07869796724825022,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 7.2137,
|
|
"mean_token_accuracy": 0.1104245975613594,
|
|
"num_tokens": 2329592.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.326994514465332,
|
|
"epoch": 0.07903285221526406,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 7.2911,
|
|
"mean_token_accuracy": 0.1015694797039032,
|
|
"num_tokens": 2339219.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.2799012660980225,
|
|
"epoch": 0.07936773718227788,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 7.2485,
|
|
"mean_token_accuracy": 0.10349665880203247,
|
|
"num_tokens": 2349037.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.291835355758667,
|
|
"epoch": 0.07970262214929172,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 7.2575,
|
|
"mean_token_accuracy": 0.10751553550362587,
|
|
"num_tokens": 2359321.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.34235315322876,
|
|
"epoch": 0.08003750711630554,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 7.3263,
|
|
"mean_token_accuracy": 0.10246394127607346,
|
|
"num_tokens": 2368978.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.249216890335083,
|
|
"epoch": 0.08037239208331938,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 7.2286,
|
|
"mean_token_accuracy": 0.10601548030972481,
|
|
"num_tokens": 2378429.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.3063520908355715,
|
|
"epoch": 0.0807072770503332,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 7.2869,
|
|
"mean_token_accuracy": 0.10153080821037293,
|
|
"num_tokens": 2389711.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.228058338165283,
|
|
"epoch": 0.08104216201734704,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 7.2833,
|
|
"mean_token_accuracy": 0.10485092476010323,
|
|
"num_tokens": 2399207.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.281473827362061,
|
|
"epoch": 0.08137704698436087,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 7.2453,
|
|
"mean_token_accuracy": 0.10985753983259201,
|
|
"num_tokens": 2409236.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.30497670173645,
|
|
"epoch": 0.0817119319513747,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 7.2938,
|
|
"mean_token_accuracy": 0.1052240014076233,
|
|
"num_tokens": 2419588.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.2253070831298825,
|
|
"epoch": 0.08204681691838854,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 7.1701,
|
|
"mean_token_accuracy": 0.10711280703544616,
|
|
"num_tokens": 2429651.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.329334735870361,
|
|
"epoch": 0.08238170188540236,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 7.3662,
|
|
"mean_token_accuracy": 0.0996719591319561,
|
|
"num_tokens": 2439942.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.277449989318848,
|
|
"epoch": 0.0827165868524162,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 7.263,
|
|
"mean_token_accuracy": 0.10270665660500526,
|
|
"num_tokens": 2448843.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.296933031082153,
|
|
"epoch": 0.08305147181943003,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 7.2922,
|
|
"mean_token_accuracy": 0.10227823108434678,
|
|
"num_tokens": 2458878.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.26516489982605,
|
|
"epoch": 0.08338635678644386,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 7.1605,
|
|
"mean_token_accuracy": 0.1101699061691761,
|
|
"num_tokens": 2468527.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.2820967674255375,
|
|
"epoch": 0.08372124175345769,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 7.3152,
|
|
"mean_token_accuracy": 0.10319520086050034,
|
|
"num_tokens": 2478996.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.2530265808105465,
|
|
"epoch": 0.08405612672047152,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 7.1887,
|
|
"mean_token_accuracy": 0.10442061051726341,
|
|
"num_tokens": 2488697.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.187231397628784,
|
|
"epoch": 0.08439101168748535,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 7.2401,
|
|
"mean_token_accuracy": 0.10387319102883338,
|
|
"num_tokens": 2498590.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.274334239959717,
|
|
"epoch": 0.08472589665449919,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 7.1824,
|
|
"mean_token_accuracy": 0.10959140658378601,
|
|
"num_tokens": 2507824.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.244437408447266,
|
|
"epoch": 0.08506078162151301,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 7.2469,
|
|
"mean_token_accuracy": 0.10558498799800872,
|
|
"num_tokens": 2516738.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.3016125679016115,
|
|
"epoch": 0.08539566658852685,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 7.2459,
|
|
"mean_token_accuracy": 0.10852355360984803,
|
|
"num_tokens": 2525843.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.2142653465271,
|
|
"epoch": 0.08573055155554067,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 7.2187,
|
|
"mean_token_accuracy": 0.10861733108758927,
|
|
"num_tokens": 2534877.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.3023340702056885,
|
|
"epoch": 0.08606543652255451,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 7.2457,
|
|
"mean_token_accuracy": 0.10270560383796692,
|
|
"num_tokens": 2545197.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.228833866119385,
|
|
"epoch": 0.08640032148956833,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 7.2251,
|
|
"mean_token_accuracy": 0.11042040884494782,
|
|
"num_tokens": 2555732.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.210680150985718,
|
|
"epoch": 0.08673520645658217,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 7.2745,
|
|
"mean_token_accuracy": 0.1047789104282856,
|
|
"num_tokens": 2565779.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.2319292545318605,
|
|
"epoch": 0.087070091423596,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 7.173,
|
|
"mean_token_accuracy": 0.11112685948610306,
|
|
"num_tokens": 2574822.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.255096387863159,
|
|
"epoch": 0.08740497639060983,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 7.2847,
|
|
"mean_token_accuracy": 0.10384105294942855,
|
|
"num_tokens": 2584780.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.21968297958374,
|
|
"epoch": 0.08773986135762366,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 7.2825,
|
|
"mean_token_accuracy": 0.10405526086688041,
|
|
"num_tokens": 2595080.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.281798458099365,
|
|
"epoch": 0.08807474632463749,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 7.3151,
|
|
"mean_token_accuracy": 0.10257799997925758,
|
|
"num_tokens": 2605221.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.314059209823609,
|
|
"epoch": 0.08840963129165132,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 7.2323,
|
|
"mean_token_accuracy": 0.10295548290014267,
|
|
"num_tokens": 2615915.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.2096466541290285,
|
|
"epoch": 0.08874451625866515,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 7.1915,
|
|
"mean_token_accuracy": 0.10639428049325943,
|
|
"num_tokens": 2626791.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.226166772842407,
|
|
"epoch": 0.08907940122567898,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 7.2656,
|
|
"mean_token_accuracy": 0.10837312117218971,
|
|
"num_tokens": 2636055.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.297124433517456,
|
|
"epoch": 0.08941428619269282,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 7.1916,
|
|
"mean_token_accuracy": 0.10522143691778182,
|
|
"num_tokens": 2645177.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.187276220321655,
|
|
"epoch": 0.08974917115970664,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 7.1588,
|
|
"mean_token_accuracy": 0.11183756738901138,
|
|
"num_tokens": 2655961.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.148924684524536,
|
|
"epoch": 0.09008405612672048,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 7.1541,
|
|
"mean_token_accuracy": 0.11793737411499024,
|
|
"num_tokens": 2666394.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.216915845870972,
|
|
"epoch": 0.0904189410937343,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 7.1927,
|
|
"mean_token_accuracy": 0.10960348248481751,
|
|
"num_tokens": 2675309.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.186840438842774,
|
|
"epoch": 0.09075382606074814,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 7.1826,
|
|
"mean_token_accuracy": 0.10617644041776657,
|
|
"num_tokens": 2685500.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.246086359024048,
|
|
"epoch": 0.09108871102776196,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 7.1035,
|
|
"mean_token_accuracy": 0.10482602342963218,
|
|
"num_tokens": 2694648.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.174089431762695,
|
|
"epoch": 0.0914235959947758,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 7.2129,
|
|
"mean_token_accuracy": 0.10653489902615547,
|
|
"num_tokens": 2703878.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.275624847412109,
|
|
"epoch": 0.09175848096178962,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 7.2574,
|
|
"mean_token_accuracy": 0.10273868069052697,
|
|
"num_tokens": 2714142.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.172191524505616,
|
|
"epoch": 0.09209336592880346,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 7.1153,
|
|
"mean_token_accuracy": 0.11495553776621818,
|
|
"num_tokens": 2723083.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.233070755004883,
|
|
"epoch": 0.09242825089581728,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 7.2819,
|
|
"mean_token_accuracy": 0.10039968714118004,
|
|
"num_tokens": 2733577.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.328128337860107,
|
|
"epoch": 0.09276313586283112,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 7.2459,
|
|
"mean_token_accuracy": 0.0923890821635723,
|
|
"num_tokens": 2744134.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.1128767967224125,
|
|
"epoch": 0.09309802082984495,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 7.1178,
|
|
"mean_token_accuracy": 0.10745147988200188,
|
|
"num_tokens": 2754460.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.22829155921936,
|
|
"epoch": 0.09343290579685878,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 7.2128,
|
|
"mean_token_accuracy": 0.10378189608454705,
|
|
"num_tokens": 2763935.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.211243963241577,
|
|
"epoch": 0.09376779076387261,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 7.1907,
|
|
"mean_token_accuracy": 0.10557325407862664,
|
|
"num_tokens": 2774945.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.171038055419922,
|
|
"epoch": 0.09410267573088645,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 7.2125,
|
|
"mean_token_accuracy": 0.10536214634776116,
|
|
"num_tokens": 2784799.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.0606297016143795,
|
|
"epoch": 0.09443756069790027,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 7.0165,
|
|
"mean_token_accuracy": 0.11608987599611283,
|
|
"num_tokens": 2794506.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.236382102966308,
|
|
"epoch": 0.0947724456649141,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 7.276,
|
|
"mean_token_accuracy": 0.1041162945330143,
|
|
"num_tokens": 2804547.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.113600969314575,
|
|
"epoch": 0.09510733063192793,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 7.1121,
|
|
"mean_token_accuracy": 0.10910582020878792,
|
|
"num_tokens": 2815279.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.247256755828857,
|
|
"epoch": 0.09544221559894177,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 7.185,
|
|
"mean_token_accuracy": 0.09964644908905029,
|
|
"num_tokens": 2825754.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.25377254486084,
|
|
"epoch": 0.09577710056595559,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 7.3647,
|
|
"mean_token_accuracy": 0.10228754505515099,
|
|
"num_tokens": 2835775.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.1915538787841795,
|
|
"epoch": 0.09611198553296943,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 7.0842,
|
|
"mean_token_accuracy": 0.10943688377737999,
|
|
"num_tokens": 2845065.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.218343782424927,
|
|
"epoch": 0.09644687049998325,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 7.2454,
|
|
"mean_token_accuracy": 0.10062536150217057,
|
|
"num_tokens": 2855493.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.202933025360108,
|
|
"epoch": 0.09678175546699709,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 7.0416,
|
|
"mean_token_accuracy": 0.11148291304707528,
|
|
"num_tokens": 2864449.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.00218358039856,
|
|
"epoch": 0.09711664043401091,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 7.0897,
|
|
"mean_token_accuracy": 0.11226318851113319,
|
|
"num_tokens": 2874056.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.200784492492676,
|
|
"epoch": 0.09745152540102475,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 7.1136,
|
|
"mean_token_accuracy": 0.10861515179276467,
|
|
"num_tokens": 2884283.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.221148538589477,
|
|
"epoch": 0.09778641036803858,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 7.1259,
|
|
"mean_token_accuracy": 0.11199686899781228,
|
|
"num_tokens": 2892720.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.0525695323944095,
|
|
"epoch": 0.09812129533505241,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 7.1113,
|
|
"mean_token_accuracy": 0.10836914032697678,
|
|
"num_tokens": 2902671.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.207837247848511,
|
|
"epoch": 0.09845618030206624,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 7.1715,
|
|
"mean_token_accuracy": 0.10252555385231972,
|
|
"num_tokens": 2912683.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.156357526779175,
|
|
"epoch": 0.09879106526908008,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 7.1258,
|
|
"mean_token_accuracy": 0.10757871195673943,
|
|
"num_tokens": 2922888.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.196430444717407,
|
|
"epoch": 0.0991259502360939,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 7.1284,
|
|
"mean_token_accuracy": 0.10644729062914848,
|
|
"num_tokens": 2932284.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.186505126953125,
|
|
"epoch": 0.09946083520310774,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 7.1535,
|
|
"mean_token_accuracy": 0.11216149106621742,
|
|
"num_tokens": 2942061.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.131209659576416,
|
|
"epoch": 0.09979572017012156,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 7.0322,
|
|
"mean_token_accuracy": 0.11146841198205948,
|
|
"num_tokens": 2950216.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.070336866378784,
|
|
"epoch": 0.1001306051371354,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 7.0925,
|
|
"mean_token_accuracy": 0.11065334677696229,
|
|
"num_tokens": 2959578.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.151112222671509,
|
|
"epoch": 0.10046549010414922,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.11227871030569077,
|
|
"num_tokens": 2969043.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"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": 4740384540672000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|