7537dbe9df
Model: fpadovani/eus-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.0068758057584875,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.90788049697876,
|
|
"epoch": 0.004297378599054577,
|
|
"grad_norm": 23.0,
|
|
"learning_rate": 2e-06,
|
|
"loss": 13.9846,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10107.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.913597154617309,
|
|
"epoch": 0.008594757198109154,
|
|
"grad_norm": 23.875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 13.7937,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 18391.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.986179208755493,
|
|
"epoch": 0.01289213579716373,
|
|
"grad_norm": 25.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.7057,
|
|
"mean_token_accuracy": 7.022471982054412e-05,
|
|
"num_tokens": 27061.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.105097055435181,
|
|
"epoch": 0.017189514396218308,
|
|
"grad_norm": 31.25,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.4704,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 36339.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.69900426864624,
|
|
"epoch": 0.021486892995272882,
|
|
"grad_norm": 38.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.6357,
|
|
"mean_token_accuracy": 0.0002869559801183641,
|
|
"num_tokens": 45770.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.67990927696228,
|
|
"epoch": 0.02578427159432746,
|
|
"grad_norm": 20.875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.573,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 54575.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.355410289764404,
|
|
"epoch": 0.030081650193382038,
|
|
"grad_norm": 3.453125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.8476,
|
|
"mean_token_accuracy": 0.0001408907352015376,
|
|
"num_tokens": 66403.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.715928554534912,
|
|
"epoch": 0.034379028792436615,
|
|
"grad_norm": 3.53125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.6808,
|
|
"mean_token_accuracy": 0.001065203151665628,
|
|
"num_tokens": 76510.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.736311149597167,
|
|
"epoch": 0.03867640739149119,
|
|
"grad_norm": 3.328125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.4612,
|
|
"mean_token_accuracy": 0.014692968456074595,
|
|
"num_tokens": 84836.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.713325881958008,
|
|
"epoch": 0.042973785990545764,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.2528,
|
|
"mean_token_accuracy": 0.021226395200937987,
|
|
"num_tokens": 93197.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.619150924682618,
|
|
"epoch": 0.047271164589600345,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.0647,
|
|
"mean_token_accuracy": 0.019917360693216323,
|
|
"num_tokens": 101546.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.670361804962159,
|
|
"epoch": 0.05156854318865492,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.9914,
|
|
"mean_token_accuracy": 0.023917235806584357,
|
|
"num_tokens": 111703.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.706762981414794,
|
|
"epoch": 0.055865921787709494,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8119,
|
|
"mean_token_accuracy": 0.05865457281470299,
|
|
"num_tokens": 119894.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.680897426605224,
|
|
"epoch": 0.060163300386764075,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7551,
|
|
"mean_token_accuracy": 0.060981281101703644,
|
|
"num_tokens": 128885.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.6582462310791,
|
|
"epoch": 0.06446067898581866,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7084,
|
|
"mean_token_accuracy": 0.05785200446844101,
|
|
"num_tokens": 138106.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.663964080810548,
|
|
"epoch": 0.06875805758487323,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.649,
|
|
"mean_token_accuracy": 0.060528215020895004,
|
|
"num_tokens": 146691.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.653559303283691,
|
|
"epoch": 0.0730554361839278,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5699,
|
|
"mean_token_accuracy": 0.060284944623708724,
|
|
"num_tokens": 155792.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.636913108825684,
|
|
"epoch": 0.07735281478298238,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5761,
|
|
"mean_token_accuracy": 0.05633340328931809,
|
|
"num_tokens": 166944.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.626116371154785,
|
|
"epoch": 0.08165019338203695,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3779,
|
|
"mean_token_accuracy": 0.06080182269215584,
|
|
"num_tokens": 175303.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.605600547790527,
|
|
"epoch": 0.08594757198109153,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3452,
|
|
"mean_token_accuracy": 0.05839457102119923,
|
|
"num_tokens": 184708.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.575385189056396,
|
|
"epoch": 0.09024495058014612,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2591,
|
|
"mean_token_accuracy": 0.06349388286471366,
|
|
"num_tokens": 193835.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.532122993469239,
|
|
"epoch": 0.09454232917920069,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1605,
|
|
"mean_token_accuracy": 0.07098993584513665,
|
|
"num_tokens": 203344.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.480678367614747,
|
|
"epoch": 0.09883970777825526,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.1509,
|
|
"mean_token_accuracy": 0.05849636979401111,
|
|
"num_tokens": 213048.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.444736099243164,
|
|
"epoch": 0.10313708637730984,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9602,
|
|
"mean_token_accuracy": 0.07322230078279972,
|
|
"num_tokens": 221784.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.375800514221192,
|
|
"epoch": 0.10743446497636441,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8772,
|
|
"mean_token_accuracy": 0.07452203780412674,
|
|
"num_tokens": 230971.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.297348403930664,
|
|
"epoch": 0.11173184357541899,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.7682,
|
|
"mean_token_accuracy": 0.07423729747533798,
|
|
"num_tokens": 240524.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.169759750366211,
|
|
"epoch": 0.11602922217447358,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6555,
|
|
"mean_token_accuracy": 0.07146260924637318,
|
|
"num_tokens": 249220.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.120108795166015,
|
|
"epoch": 0.12032660077352815,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6406,
|
|
"mean_token_accuracy": 0.0681857068091631,
|
|
"num_tokens": 258934.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.981133079528808,
|
|
"epoch": 0.12462397937258272,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.5023,
|
|
"mean_token_accuracy": 0.06928090676665306,
|
|
"num_tokens": 267680.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.886031436920167,
|
|
"epoch": 0.1289213579716373,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.4198,
|
|
"mean_token_accuracy": 0.07501428201794624,
|
|
"num_tokens": 276227.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.753608226776123,
|
|
"epoch": 0.1332187365706919,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.2874,
|
|
"mean_token_accuracy": 0.07755625247955322,
|
|
"num_tokens": 286342.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.609795665740966,
|
|
"epoch": 0.13751611516974646,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.2601,
|
|
"mean_token_accuracy": 0.06969402097165585,
|
|
"num_tokens": 294994.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.492193508148194,
|
|
"epoch": 0.14181349376880104,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.1727,
|
|
"mean_token_accuracy": 0.07241848222911358,
|
|
"num_tokens": 303882.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.254251098632812,
|
|
"epoch": 0.1461108723678556,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.0965,
|
|
"mean_token_accuracy": 0.07132276222109794,
|
|
"num_tokens": 312515.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.1218976020813,
|
|
"epoch": 0.15040825096691018,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.0532,
|
|
"mean_token_accuracy": 0.06978616863489151,
|
|
"num_tokens": 320801.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.900997734069824,
|
|
"epoch": 0.15470562956596476,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.0224,
|
|
"mean_token_accuracy": 0.07444634772837162,
|
|
"num_tokens": 329382.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.709702205657958,
|
|
"epoch": 0.15900300816501933,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.9836,
|
|
"mean_token_accuracy": 0.06757781356573105,
|
|
"num_tokens": 337894.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.654694271087646,
|
|
"epoch": 0.1633003867640739,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.9562,
|
|
"mean_token_accuracy": 0.06648631133139134,
|
|
"num_tokens": 346380.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.521081924438477,
|
|
"epoch": 0.16759776536312848,
|
|
"grad_norm": 0.65625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.9991,
|
|
"mean_token_accuracy": 0.06590078994631768,
|
|
"num_tokens": 356305.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.442760276794434,
|
|
"epoch": 0.17189514396218306,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.9449,
|
|
"mean_token_accuracy": 0.06936337985098362,
|
|
"num_tokens": 364899.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.41897268295288,
|
|
"epoch": 0.17619252256123766,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.9196,
|
|
"mean_token_accuracy": 0.0691080518066883,
|
|
"num_tokens": 373663.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.352293586730957,
|
|
"epoch": 0.18048990116029223,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.8931,
|
|
"mean_token_accuracy": 0.07045749872922898,
|
|
"num_tokens": 382730.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.23891839981079,
|
|
"epoch": 0.1847872797593468,
|
|
"grad_norm": 0.67578125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.9378,
|
|
"mean_token_accuracy": 0.06828119307756424,
|
|
"num_tokens": 392676.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.277576446533203,
|
|
"epoch": 0.18908465835840138,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.9179,
|
|
"mean_token_accuracy": 0.07761286348104476,
|
|
"num_tokens": 401050.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.313580322265626,
|
|
"epoch": 0.19338203695745596,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.8686,
|
|
"mean_token_accuracy": 0.0736297857016325,
|
|
"num_tokens": 410009.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.305255031585693,
|
|
"epoch": 0.19767941555651053,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.8585,
|
|
"mean_token_accuracy": 0.07423750944435596,
|
|
"num_tokens": 419302.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.293313884735108,
|
|
"epoch": 0.2019767941555651,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7711,
|
|
"mean_token_accuracy": 0.08085694760084153,
|
|
"num_tokens": 427296.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.249619960784912,
|
|
"epoch": 0.20627417275461968,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.8993,
|
|
"mean_token_accuracy": 0.07239912934601307,
|
|
"num_tokens": 436368.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.27509069442749,
|
|
"epoch": 0.21057155135367425,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.8229,
|
|
"mean_token_accuracy": 0.0778175387531519,
|
|
"num_tokens": 445535.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.245413112640382,
|
|
"epoch": 0.21486892995272883,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.7971,
|
|
"mean_token_accuracy": 0.07436932176351548,
|
|
"num_tokens": 454769.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.240999603271485,
|
|
"epoch": 0.2191663085517834,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.8138,
|
|
"mean_token_accuracy": 0.07600181177258492,
|
|
"num_tokens": 463975.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.211737537384034,
|
|
"epoch": 0.22346368715083798,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.8027,
|
|
"mean_token_accuracy": 0.08170078098773956,
|
|
"num_tokens": 472899.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.172839450836182,
|
|
"epoch": 0.22776106574989258,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.8171,
|
|
"mean_token_accuracy": 0.07734721302986144,
|
|
"num_tokens": 481556.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.150885677337646,
|
|
"epoch": 0.23205844434894715,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.7804,
|
|
"mean_token_accuracy": 0.07963913902640343,
|
|
"num_tokens": 490253.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.222941017150879,
|
|
"epoch": 0.23635582294800173,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.785,
|
|
"mean_token_accuracy": 0.07519972808659077,
|
|
"num_tokens": 498444.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.142861270904541,
|
|
"epoch": 0.2406532015470563,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.7267,
|
|
"mean_token_accuracy": 0.08277813419699669,
|
|
"num_tokens": 508330.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.140101432800293,
|
|
"epoch": 0.24495058014611087,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.8112,
|
|
"mean_token_accuracy": 0.0772799663245678,
|
|
"num_tokens": 517900.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.134088563919068,
|
|
"epoch": 0.24924795874516545,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.7764,
|
|
"mean_token_accuracy": 0.0777011513710022,
|
|
"num_tokens": 527808.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.083810663223266,
|
|
"epoch": 0.25354533734422,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.7652,
|
|
"mean_token_accuracy": 0.07280438467860222,
|
|
"num_tokens": 536931.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.169271469116211,
|
|
"epoch": 0.2578427159432746,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.7957,
|
|
"mean_token_accuracy": 0.0796648234128952,
|
|
"num_tokens": 545758.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.125834083557129,
|
|
"epoch": 0.26214009454232917,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.7996,
|
|
"mean_token_accuracy": 0.07962435148656369,
|
|
"num_tokens": 555165.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.13117265701294,
|
|
"epoch": 0.2664374731413838,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.7089,
|
|
"mean_token_accuracy": 0.07958239540457726,
|
|
"num_tokens": 564719.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.057837867736817,
|
|
"epoch": 0.2707348517404383,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6928,
|
|
"mean_token_accuracy": 0.0828514739871025,
|
|
"num_tokens": 573572.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.223217296600343,
|
|
"epoch": 0.2750322303394929,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.8692,
|
|
"mean_token_accuracy": 0.07304716818034648,
|
|
"num_tokens": 581497.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.053431844711303,
|
|
"epoch": 0.27932960893854747,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.7017,
|
|
"mean_token_accuracy": 0.08142868839204312,
|
|
"num_tokens": 591107.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.063822364807129,
|
|
"epoch": 0.28362698753760207,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.7406,
|
|
"mean_token_accuracy": 0.0793286319822073,
|
|
"num_tokens": 600241.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.13500518798828,
|
|
"epoch": 0.2879243661366566,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.6735,
|
|
"mean_token_accuracy": 0.07961011826992034,
|
|
"num_tokens": 608697.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.040066146850586,
|
|
"epoch": 0.2922217447357112,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.6584,
|
|
"mean_token_accuracy": 0.08232032433152199,
|
|
"num_tokens": 617275.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.135028553009032,
|
|
"epoch": 0.29651912333476577,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.7489,
|
|
"mean_token_accuracy": 0.08271522969007492,
|
|
"num_tokens": 626644.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.011529541015625,
|
|
"epoch": 0.30081650193382037,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.6696,
|
|
"mean_token_accuracy": 0.0852140061557293,
|
|
"num_tokens": 635110.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.075648546218872,
|
|
"epoch": 0.30511388053287497,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.8072,
|
|
"mean_token_accuracy": 0.07943284660577773,
|
|
"num_tokens": 644746.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.112467098236085,
|
|
"epoch": 0.3094112591319295,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.6965,
|
|
"mean_token_accuracy": 0.08117582276463509,
|
|
"num_tokens": 654281.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.078112173080445,
|
|
"epoch": 0.3137086377309841,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.6883,
|
|
"mean_token_accuracy": 0.08609185442328453,
|
|
"num_tokens": 663174.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.986557722091675,
|
|
"epoch": 0.31800601633003867,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.6259,
|
|
"mean_token_accuracy": 0.08559975251555443,
|
|
"num_tokens": 672178.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.04137225151062,
|
|
"epoch": 0.32230339492909327,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.6371,
|
|
"mean_token_accuracy": 0.08901203349232674,
|
|
"num_tokens": 681323.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.031937456130981,
|
|
"epoch": 0.3266007735281478,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.5447,
|
|
"mean_token_accuracy": 0.08979653343558311,
|
|
"num_tokens": 690461.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.969046926498413,
|
|
"epoch": 0.3308981521272024,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.61,
|
|
"mean_token_accuracy": 0.07953116819262504,
|
|
"num_tokens": 699199.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.007894515991211,
|
|
"epoch": 0.33519553072625696,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.5729,
|
|
"mean_token_accuracy": 0.08715594932436943,
|
|
"num_tokens": 707949.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.037422609329223,
|
|
"epoch": 0.33949290932531156,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.6302,
|
|
"mean_token_accuracy": 0.08520551696419716,
|
|
"num_tokens": 715752.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.013258504867554,
|
|
"epoch": 0.3437902879243661,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.5717,
|
|
"mean_token_accuracy": 0.08992563635110855,
|
|
"num_tokens": 724416.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.055727624893189,
|
|
"epoch": 0.3480876665234207,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.5987,
|
|
"mean_token_accuracy": 0.08299638405442238,
|
|
"num_tokens": 733116.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.958417224884033,
|
|
"epoch": 0.3523850451224753,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5847,
|
|
"mean_token_accuracy": 0.08687192872166634,
|
|
"num_tokens": 742093.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.976228284835815,
|
|
"epoch": 0.35668242372152986,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.5894,
|
|
"mean_token_accuracy": 0.0836369551718235,
|
|
"num_tokens": 750402.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.9206877708435055,
|
|
"epoch": 0.36097980232058446,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5472,
|
|
"mean_token_accuracy": 0.08791278377175331,
|
|
"num_tokens": 760961.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.975084543228149,
|
|
"epoch": 0.365277180919639,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.6033,
|
|
"mean_token_accuracy": 0.08895924240350724,
|
|
"num_tokens": 770554.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.908215951919556,
|
|
"epoch": 0.3695745595186936,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.5822,
|
|
"mean_token_accuracy": 0.08788935765624047,
|
|
"num_tokens": 779172.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.003646516799927,
|
|
"epoch": 0.37387193811774816,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.6269,
|
|
"mean_token_accuracy": 0.0811423234641552,
|
|
"num_tokens": 788040.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.020764684677124,
|
|
"epoch": 0.37816931671680276,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.5218,
|
|
"mean_token_accuracy": 0.08682032898068429,
|
|
"num_tokens": 796786.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.909510374069214,
|
|
"epoch": 0.3824666953158573,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.55,
|
|
"mean_token_accuracy": 0.08918910399079323,
|
|
"num_tokens": 805520.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.937164402008056,
|
|
"epoch": 0.3867640739149119,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.4594,
|
|
"mean_token_accuracy": 0.08978109285235406,
|
|
"num_tokens": 814939.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.91705961227417,
|
|
"epoch": 0.39106145251396646,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.4827,
|
|
"mean_token_accuracy": 0.08797861486673356,
|
|
"num_tokens": 823862.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.884686851501465,
|
|
"epoch": 0.39535883111302106,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.499,
|
|
"mean_token_accuracy": 0.08911096453666686,
|
|
"num_tokens": 832820.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.961242198944092,
|
|
"epoch": 0.39965620971207566,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.463,
|
|
"mean_token_accuracy": 0.08996951058506966,
|
|
"num_tokens": 841538.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.896987247467041,
|
|
"epoch": 0.4039535883111302,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.5646,
|
|
"mean_token_accuracy": 0.0861456498503685,
|
|
"num_tokens": 851123.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.01444125175476,
|
|
"epoch": 0.4082509669101848,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.5439,
|
|
"mean_token_accuracy": 0.0893861047923565,
|
|
"num_tokens": 860357.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.83814754486084,
|
|
"epoch": 0.41254834550923936,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.458,
|
|
"mean_token_accuracy": 0.09221402034163476,
|
|
"num_tokens": 869980.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.830886650085449,
|
|
"epoch": 0.41684572410829396,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.4055,
|
|
"mean_token_accuracy": 0.08723795711994171,
|
|
"num_tokens": 878250.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.888704490661621,
|
|
"epoch": 0.4211431027073485,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.5116,
|
|
"mean_token_accuracy": 0.08683100938796998,
|
|
"num_tokens": 887624.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.814441156387329,
|
|
"epoch": 0.4254404813064031,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.4596,
|
|
"mean_token_accuracy": 0.08961054012179374,
|
|
"num_tokens": 897120.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.8445289611816404,
|
|
"epoch": 0.42973785990545765,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.4306,
|
|
"mean_token_accuracy": 0.09552830010652542,
|
|
"num_tokens": 906215.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.42973785990545765,
|
|
"eval_entropy": 7.63128979463835,
|
|
"eval_loss": 7.479809284210205,
|
|
"eval_mean_token_accuracy": 0.0912898702228123,
|
|
"eval_num_tokens": 906215.0,
|
|
"eval_runtime": 2.0631,
|
|
"eval_samples_per_second": 1720.253,
|
|
"eval_steps_per_second": 215.213,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.809079313278199,
|
|
"epoch": 0.43403523850451226,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.4118,
|
|
"mean_token_accuracy": 0.08965951427817345,
|
|
"num_tokens": 915181.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.847093725204468,
|
|
"epoch": 0.4383326171035668,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.429,
|
|
"mean_token_accuracy": 0.09137469455599785,
|
|
"num_tokens": 924377.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.866995763778687,
|
|
"epoch": 0.4426299957026214,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.5053,
|
|
"mean_token_accuracy": 0.09116312190890312,
|
|
"num_tokens": 933114.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.865507459640503,
|
|
"epoch": 0.44692737430167595,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.4689,
|
|
"mean_token_accuracy": 0.08555961102247238,
|
|
"num_tokens": 943306.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.781367206573487,
|
|
"epoch": 0.45122475290073055,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.3577,
|
|
"mean_token_accuracy": 0.09929761290550232,
|
|
"num_tokens": 951515.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.766856384277344,
|
|
"epoch": 0.45552213149978515,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.4694,
|
|
"mean_token_accuracy": 0.09030876457691192,
|
|
"num_tokens": 962686.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.901469707489014,
|
|
"epoch": 0.4598195100988397,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.4753,
|
|
"mean_token_accuracy": 0.09142492190003396,
|
|
"num_tokens": 972136.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.716486024856567,
|
|
"epoch": 0.4641168886978943,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.3638,
|
|
"mean_token_accuracy": 0.09268842414021491,
|
|
"num_tokens": 981301.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.767406940460205,
|
|
"epoch": 0.46841426729694885,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.3949,
|
|
"mean_token_accuracy": 0.09462741911411285,
|
|
"num_tokens": 990360.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.815031290054321,
|
|
"epoch": 0.47271164589600345,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.4369,
|
|
"mean_token_accuracy": 0.09093201607465744,
|
|
"num_tokens": 999415.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.7446294784545895,
|
|
"epoch": 0.477009024495058,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.3662,
|
|
"mean_token_accuracy": 0.09483416303992272,
|
|
"num_tokens": 1008762.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.768563890457154,
|
|
"epoch": 0.4813064030941126,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.3624,
|
|
"mean_token_accuracy": 0.09757203981280327,
|
|
"num_tokens": 1017704.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.691549444198609,
|
|
"epoch": 0.48560378169316715,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.3155,
|
|
"mean_token_accuracy": 0.0982676163315773,
|
|
"num_tokens": 1026251.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.725941944122314,
|
|
"epoch": 0.48990116029222175,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.2996,
|
|
"mean_token_accuracy": 0.09966658651828766,
|
|
"num_tokens": 1036191.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.717142486572266,
|
|
"epoch": 0.4941985388912763,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.3328,
|
|
"mean_token_accuracy": 0.09679939597845078,
|
|
"num_tokens": 1044936.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.726046752929688,
|
|
"epoch": 0.4984959174903309,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.3382,
|
|
"mean_token_accuracy": 0.09426090195775032,
|
|
"num_tokens": 1053683.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.633957862854004,
|
|
"epoch": 0.5027932960893855,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.3198,
|
|
"mean_token_accuracy": 0.09904149472713471,
|
|
"num_tokens": 1062932.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.770825815200806,
|
|
"epoch": 0.50709067468844,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.3476,
|
|
"mean_token_accuracy": 0.09258887320756912,
|
|
"num_tokens": 1072313.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.691164636611939,
|
|
"epoch": 0.5113880532874946,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.3104,
|
|
"mean_token_accuracy": 0.09508624523878098,
|
|
"num_tokens": 1081675.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.71315016746521,
|
|
"epoch": 0.5156854318865493,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.3272,
|
|
"mean_token_accuracy": 0.09484435245394707,
|
|
"num_tokens": 1091541.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.773202800750733,
|
|
"epoch": 0.5199828104856038,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.3362,
|
|
"mean_token_accuracy": 0.10049614533782006,
|
|
"num_tokens": 1100724.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.529764127731323,
|
|
"epoch": 0.5242801890846583,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.2799,
|
|
"mean_token_accuracy": 0.10042180344462395,
|
|
"num_tokens": 1108869.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.751347589492798,
|
|
"epoch": 0.5285775676837129,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.3109,
|
|
"mean_token_accuracy": 0.09996724799275399,
|
|
"num_tokens": 1117314.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.551737356185913,
|
|
"epoch": 0.5328749462827675,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.2409,
|
|
"mean_token_accuracy": 0.09862213358283042,
|
|
"num_tokens": 1126786.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.609012699127197,
|
|
"epoch": 0.5371723248818221,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.2349,
|
|
"mean_token_accuracy": 0.09517790451645851,
|
|
"num_tokens": 1136013.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.7104791641235355,
|
|
"epoch": 0.5414697034808766,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.2709,
|
|
"mean_token_accuracy": 0.10074363201856613,
|
|
"num_tokens": 1144970.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.587553977966309,
|
|
"epoch": 0.5457670820799312,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.2408,
|
|
"mean_token_accuracy": 0.1027376800775528,
|
|
"num_tokens": 1153810.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.597120857238769,
|
|
"epoch": 0.5500644606789858,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.2171,
|
|
"mean_token_accuracy": 0.10400616675615311,
|
|
"num_tokens": 1162498.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.627674579620361,
|
|
"epoch": 0.5543618392780404,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.3008,
|
|
"mean_token_accuracy": 0.09583014771342277,
|
|
"num_tokens": 1172091.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.639283657073975,
|
|
"epoch": 0.5586592178770949,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.2766,
|
|
"mean_token_accuracy": 0.09886594414710999,
|
|
"num_tokens": 1181400.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.654122257232666,
|
|
"epoch": 0.5629565964761496,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.2462,
|
|
"mean_token_accuracy": 0.10131431221961976,
|
|
"num_tokens": 1189780.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.497343397140503,
|
|
"epoch": 0.5672539750752041,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.2029,
|
|
"mean_token_accuracy": 0.10030499249696731,
|
|
"num_tokens": 1198671.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.620815324783325,
|
|
"epoch": 0.5715513536742587,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.2489,
|
|
"mean_token_accuracy": 0.0975674532353878,
|
|
"num_tokens": 1207173.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.556894588470459,
|
|
"epoch": 0.5758487322733132,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.1783,
|
|
"mean_token_accuracy": 0.10579388439655305,
|
|
"num_tokens": 1216387.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.638325643539429,
|
|
"epoch": 0.5801461108723679,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.1487,
|
|
"mean_token_accuracy": 0.10557905286550522,
|
|
"num_tokens": 1224461.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.557331895828247,
|
|
"epoch": 0.5844434894714224,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.2067,
|
|
"mean_token_accuracy": 0.10470035746693611,
|
|
"num_tokens": 1233774.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.48933973312378,
|
|
"epoch": 0.588740868070477,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.173,
|
|
"mean_token_accuracy": 0.1077287070453167,
|
|
"num_tokens": 1242812.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.698177099227905,
|
|
"epoch": 0.5930382466695315,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.2652,
|
|
"mean_token_accuracy": 0.09988318458199501,
|
|
"num_tokens": 1252872.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.376710081100464,
|
|
"epoch": 0.5973356252685862,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.0588,
|
|
"mean_token_accuracy": 0.10656947568058968,
|
|
"num_tokens": 1260852.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.442113256454467,
|
|
"epoch": 0.6016330038676407,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.1521,
|
|
"mean_token_accuracy": 0.10797703564167023,
|
|
"num_tokens": 1268925.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.575676298141479,
|
|
"epoch": 0.6059303824666953,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.2053,
|
|
"mean_token_accuracy": 0.1014741227030754,
|
|
"num_tokens": 1278994.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.616156387329101,
|
|
"epoch": 0.6102277610657499,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.1986,
|
|
"mean_token_accuracy": 0.1000370778143406,
|
|
"num_tokens": 1287698.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.500000095367431,
|
|
"epoch": 0.6145251396648045,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.162,
|
|
"mean_token_accuracy": 0.10414053946733474,
|
|
"num_tokens": 1297475.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.507882452011108,
|
|
"epoch": 0.618822518263859,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.1852,
|
|
"mean_token_accuracy": 0.10390809029340745,
|
|
"num_tokens": 1306836.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.507648563385009,
|
|
"epoch": 0.6231198968629136,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.2148,
|
|
"mean_token_accuracy": 0.10065656825900078,
|
|
"num_tokens": 1315872.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.490116739273072,
|
|
"epoch": 0.6274172754619682,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.163,
|
|
"mean_token_accuracy": 0.10862415507435799,
|
|
"num_tokens": 1324624.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.495247173309326,
|
|
"epoch": 0.6317146540610228,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.1357,
|
|
"mean_token_accuracy": 0.11479571759700775,
|
|
"num_tokens": 1333058.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.333670282363892,
|
|
"epoch": 0.6360120326600773,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.0924,
|
|
"mean_token_accuracy": 0.10395059809088707,
|
|
"num_tokens": 1342376.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.5272722244262695,
|
|
"epoch": 0.6403094112591319,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.1365,
|
|
"mean_token_accuracy": 0.09796565026044846,
|
|
"num_tokens": 1351386.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.43572735786438,
|
|
"epoch": 0.6446067898581865,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.0538,
|
|
"mean_token_accuracy": 0.11305336132645608,
|
|
"num_tokens": 1358958.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.471644115447998,
|
|
"epoch": 0.6489041684572411,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.1846,
|
|
"mean_token_accuracy": 0.10423881933093071,
|
|
"num_tokens": 1368599.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.529570055007935,
|
|
"epoch": 0.6532015470562956,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.2099,
|
|
"mean_token_accuracy": 0.10423211976885796,
|
|
"num_tokens": 1378529.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.447458553314209,
|
|
"epoch": 0.6574989256553503,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.995,
|
|
"mean_token_accuracy": 0.11015288382768632,
|
|
"num_tokens": 1386993.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.396367502212525,
|
|
"epoch": 0.6617963042544048,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.0909,
|
|
"mean_token_accuracy": 0.11230141893029214,
|
|
"num_tokens": 1395790.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.3504656791687015,
|
|
"epoch": 0.6660936828534594,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.1044,
|
|
"mean_token_accuracy": 0.11002744212746621,
|
|
"num_tokens": 1405587.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.394146966934204,
|
|
"epoch": 0.6703910614525139,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.0762,
|
|
"mean_token_accuracy": 0.11049489751458168,
|
|
"num_tokens": 1414478.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.4956427097320555,
|
|
"epoch": 0.6746884400515686,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.1402,
|
|
"mean_token_accuracy": 0.11158580034971237,
|
|
"num_tokens": 1423791.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.476742601394653,
|
|
"epoch": 0.6789858186506231,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.1666,
|
|
"mean_token_accuracy": 0.10097377747297287,
|
|
"num_tokens": 1432955.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.437243223190308,
|
|
"epoch": 0.6832831972496777,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.142,
|
|
"mean_token_accuracy": 0.10785873606801033,
|
|
"num_tokens": 1441907.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.319718980789185,
|
|
"epoch": 0.6875805758487322,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.009,
|
|
"mean_token_accuracy": 0.11284941285848618,
|
|
"num_tokens": 1451062.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.428938245773315,
|
|
"epoch": 0.6918779544477869,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.0856,
|
|
"mean_token_accuracy": 0.1089414782822132,
|
|
"num_tokens": 1460132.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.379511928558349,
|
|
"epoch": 0.6961753330468414,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.9721,
|
|
"mean_token_accuracy": 0.11648677214980126,
|
|
"num_tokens": 1469582.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.305799245834351,
|
|
"epoch": 0.700472711645896,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.0538,
|
|
"mean_token_accuracy": 0.10682568848133087,
|
|
"num_tokens": 1479053.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.325813055038452,
|
|
"epoch": 0.7047700902449506,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.0643,
|
|
"mean_token_accuracy": 0.10892463400959969,
|
|
"num_tokens": 1488189.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.319435358047485,
|
|
"epoch": 0.7090674688440052,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.9276,
|
|
"mean_token_accuracy": 0.11442206054925919,
|
|
"num_tokens": 1497324.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.347770023345947,
|
|
"epoch": 0.7133648474430597,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.0315,
|
|
"mean_token_accuracy": 0.11255533620715141,
|
|
"num_tokens": 1506543.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.312428903579712,
|
|
"epoch": 0.7176622260421143,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.022,
|
|
"mean_token_accuracy": 0.11886205449700356,
|
|
"num_tokens": 1516737.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.217544221878052,
|
|
"epoch": 0.7219596046411689,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.8992,
|
|
"mean_token_accuracy": 0.1128052294254303,
|
|
"num_tokens": 1525561.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.309549617767334,
|
|
"epoch": 0.7262569832402235,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.9283,
|
|
"mean_token_accuracy": 0.11749697029590607,
|
|
"num_tokens": 1533323.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.349371337890625,
|
|
"epoch": 0.730554361839278,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.9859,
|
|
"mean_token_accuracy": 0.11753521487116814,
|
|
"num_tokens": 1542632.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.286138200759888,
|
|
"epoch": 0.7348517404383326,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.9421,
|
|
"mean_token_accuracy": 0.11514310464262963,
|
|
"num_tokens": 1551236.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.193848180770874,
|
|
"epoch": 0.7391491190373872,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.9406,
|
|
"mean_token_accuracy": 0.11694173738360406,
|
|
"num_tokens": 1559674.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.279930353164673,
|
|
"epoch": 0.7434464976364418,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.0219,
|
|
"mean_token_accuracy": 0.11416142880916595,
|
|
"num_tokens": 1569481.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.322668170928955,
|
|
"epoch": 0.7477438762354963,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.0371,
|
|
"mean_token_accuracy": 0.11332832500338555,
|
|
"num_tokens": 1578488.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.236435031890869,
|
|
"epoch": 0.752041254834551,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.9403,
|
|
"mean_token_accuracy": 0.11505360826849938,
|
|
"num_tokens": 1586675.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.316900205612183,
|
|
"epoch": 0.7563386334336055,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.048,
|
|
"mean_token_accuracy": 0.10665819644927979,
|
|
"num_tokens": 1595411.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.344276475906372,
|
|
"epoch": 0.7606360120326601,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.0296,
|
|
"mean_token_accuracy": 0.11478703245520591,
|
|
"num_tokens": 1604046.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.300757169723511,
|
|
"epoch": 0.7649333906317146,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.0139,
|
|
"mean_token_accuracy": 0.10667152628302574,
|
|
"num_tokens": 1613759.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.2238458633422855,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.9705,
|
|
"mean_token_accuracy": 0.11665266454219818,
|
|
"num_tokens": 1623323.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.301079988479614,
|
|
"epoch": 0.7735281478298238,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.9563,
|
|
"mean_token_accuracy": 0.11468923836946487,
|
|
"num_tokens": 1631727.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.2134482860565186,
|
|
"epoch": 0.7778255264288784,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.9011,
|
|
"mean_token_accuracy": 0.11652599200606346,
|
|
"num_tokens": 1639544.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.3395082473754885,
|
|
"epoch": 0.7821229050279329,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.0207,
|
|
"mean_token_accuracy": 0.10909055322408676,
|
|
"num_tokens": 1648931.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.182226657867432,
|
|
"epoch": 0.7864202836269876,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.8925,
|
|
"mean_token_accuracy": 0.11583453565835952,
|
|
"num_tokens": 1657688.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.263977575302124,
|
|
"epoch": 0.7907176622260421,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.961,
|
|
"mean_token_accuracy": 0.11562951728701591,
|
|
"num_tokens": 1666879.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.198910236358643,
|
|
"epoch": 0.7950150408250967,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.9431,
|
|
"mean_token_accuracy": 0.11777721717953682,
|
|
"num_tokens": 1676773.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.280461740493775,
|
|
"epoch": 0.7993124194241513,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.0482,
|
|
"mean_token_accuracy": 0.10834663063287735,
|
|
"num_tokens": 1686144.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.22996711730957,
|
|
"epoch": 0.8036097980232059,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.019,
|
|
"mean_token_accuracy": 0.10943668559193612,
|
|
"num_tokens": 1695476.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.193253135681152,
|
|
"epoch": 0.8079071766222604,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.9077,
|
|
"mean_token_accuracy": 0.11359465345740319,
|
|
"num_tokens": 1704907.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.2548192024230955,
|
|
"epoch": 0.812204555221315,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.9424,
|
|
"mean_token_accuracy": 0.11716259941458702,
|
|
"num_tokens": 1714564.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.272028970718384,
|
|
"epoch": 0.8165019338203696,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.9794,
|
|
"mean_token_accuracy": 0.11456828936934471,
|
|
"num_tokens": 1725285.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.167596483230591,
|
|
"epoch": 0.8207993124194242,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.9522,
|
|
"mean_token_accuracy": 0.11233177706599236,
|
|
"num_tokens": 1734331.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.214958715438843,
|
|
"epoch": 0.8250966910184787,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8938,
|
|
"mean_token_accuracy": 0.11346546709537506,
|
|
"num_tokens": 1742340.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.167982864379883,
|
|
"epoch": 0.8293940696175333,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8975,
|
|
"mean_token_accuracy": 0.11846677288413048,
|
|
"num_tokens": 1751725.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.095029211044311,
|
|
"epoch": 0.8336914482165879,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.8141,
|
|
"mean_token_accuracy": 0.12262827306985855,
|
|
"num_tokens": 1760294.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.099491214752197,
|
|
"epoch": 0.8379888268156425,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.8854,
|
|
"mean_token_accuracy": 0.11621206700801849,
|
|
"num_tokens": 1768912.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.235454845428467,
|
|
"epoch": 0.842286205414697,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.0682,
|
|
"mean_token_accuracy": 0.11861302182078362,
|
|
"num_tokens": 1778633.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.143100643157959,
|
|
"epoch": 0.8465835840137517,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.9304,
|
|
"mean_token_accuracy": 0.11692859306931495,
|
|
"num_tokens": 1787275.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.2316286087036135,
|
|
"epoch": 0.8508809626128062,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.9348,
|
|
"mean_token_accuracy": 0.11131151169538497,
|
|
"num_tokens": 1795994.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.094430828094483,
|
|
"epoch": 0.8551783412118608,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.8575,
|
|
"mean_token_accuracy": 0.11684277504682541,
|
|
"num_tokens": 1806379.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.199878978729248,
|
|
"epoch": 0.8594757198109153,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.878,
|
|
"mean_token_accuracy": 0.1225502647459507,
|
|
"num_tokens": 1816135.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8594757198109153,
|
|
"eval_entropy": 6.741199510591525,
|
|
"eval_loss": 6.911599636077881,
|
|
"eval_mean_token_accuracy": 0.11959127177324919,
|
|
"eval_num_tokens": 1816135.0,
|
|
"eval_runtime": 2.0596,
|
|
"eval_samples_per_second": 1723.145,
|
|
"eval_steps_per_second": 215.575,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.135749387741089,
|
|
"epoch": 0.86377309840997,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004999998427807679,
|
|
"loss": 6.88,
|
|
"mean_token_accuracy": 0.11578838974237442,
|
|
"num_tokens": 1824777.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.0467671871185305,
|
|
"epoch": 0.8680704770090245,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999992040780138,
|
|
"loss": 6.9483,
|
|
"mean_token_accuracy": 0.11809242218732834,
|
|
"num_tokens": 1833807.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.2159933090209964,
|
|
"epoch": 0.8723678556080791,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999980740669294,
|
|
"loss": 6.8907,
|
|
"mean_token_accuracy": 0.11748233065009117,
|
|
"num_tokens": 1843375.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.1540666103363035,
|
|
"epoch": 0.8766652342071336,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999964527499823,
|
|
"loss": 6.948,
|
|
"mean_token_accuracy": 0.11102593019604683,
|
|
"num_tokens": 1853036.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.12061505317688,
|
|
"epoch": 0.8809626128061883,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999943401307127,
|
|
"loss": 6.9196,
|
|
"mean_token_accuracy": 0.11150210499763488,
|
|
"num_tokens": 1862041.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.122325229644775,
|
|
"epoch": 0.8852599914052428,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999917362137337,
|
|
"loss": 6.8248,
|
|
"mean_token_accuracy": 0.12204656526446342,
|
|
"num_tokens": 1870707.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.1019820213317875,
|
|
"epoch": 0.8895573700042974,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999886410047312,
|
|
"loss": 6.815,
|
|
"mean_token_accuracy": 0.12252610921859741,
|
|
"num_tokens": 1879787.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.131076002120972,
|
|
"epoch": 0.8938547486033519,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999850545104638,
|
|
"loss": 6.8854,
|
|
"mean_token_accuracy": 0.11829843446612358,
|
|
"num_tokens": 1889413.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.105505514144897,
|
|
"epoch": 0.8981521272024066,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999809767387633,
|
|
"loss": 6.8543,
|
|
"mean_token_accuracy": 0.11870720088481904,
|
|
"num_tokens": 1898283.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.1863609790802006,
|
|
"epoch": 0.9024495058014611,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999764076985337,
|
|
"loss": 6.8726,
|
|
"mean_token_accuracy": 0.11925134509801864,
|
|
"num_tokens": 1907175.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.012548351287842,
|
|
"epoch": 0.9067468844005157,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999713473997519,
|
|
"loss": 6.9285,
|
|
"mean_token_accuracy": 0.11801080331206322,
|
|
"num_tokens": 1918223.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.097743034362793,
|
|
"epoch": 0.9110442629995703,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999657958534677,
|
|
"loss": 6.8618,
|
|
"mean_token_accuracy": 0.11624032855033875,
|
|
"num_tokens": 1928801.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.046412801742553,
|
|
"epoch": 0.9153416415986249,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999597530718034,
|
|
"loss": 6.8171,
|
|
"mean_token_accuracy": 0.12201576456427574,
|
|
"num_tokens": 1937406.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.076975727081299,
|
|
"epoch": 0.9196390201976794,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499953219067954,
|
|
"loss": 6.829,
|
|
"mean_token_accuracy": 0.11706628054380416,
|
|
"num_tokens": 1947184.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.0976677417755125,
|
|
"epoch": 0.923936398796734,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999461938561873,
|
|
"loss": 6.8518,
|
|
"mean_token_accuracy": 0.12058016434311866,
|
|
"num_tokens": 1956293.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.072698259353638,
|
|
"epoch": 0.9282337773957886,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999386774518432,
|
|
"loss": 6.8318,
|
|
"mean_token_accuracy": 0.11745714917778968,
|
|
"num_tokens": 1964791.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.01435604095459,
|
|
"epoch": 0.9325311559948432,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999306698713349,
|
|
"loss": 6.7357,
|
|
"mean_token_accuracy": 0.12115372568368912,
|
|
"num_tokens": 1973754.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.059027576446534,
|
|
"epoch": 0.9368285345938977,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999221711321477,
|
|
"loss": 6.8225,
|
|
"mean_token_accuracy": 0.12251915633678437,
|
|
"num_tokens": 1983035.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.931616544723511,
|
|
"epoch": 0.9411259131929522,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999131812528393,
|
|
"loss": 6.8285,
|
|
"mean_token_accuracy": 0.11646857559680938,
|
|
"num_tokens": 1992584.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.093084859848022,
|
|
"epoch": 0.9454232917920069,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00049990370025304,
|
|
"loss": 6.8713,
|
|
"mean_token_accuracy": 0.11829434335231781,
|
|
"num_tokens": 2001876.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.065115928649902,
|
|
"epoch": 0.9497206703910615,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998937281534526,
|
|
"loss": 6.7715,
|
|
"mean_token_accuracy": 0.12478053271770477,
|
|
"num_tokens": 2011067.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.09222378730774,
|
|
"epoch": 0.954018048990116,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998832649758521,
|
|
"loss": 6.847,
|
|
"mean_token_accuracy": 0.11962009072303773,
|
|
"num_tokens": 2020763.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.995938444137574,
|
|
"epoch": 0.9583154275891707,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004998723107430862,
|
|
"loss": 6.8303,
|
|
"mean_token_accuracy": 0.1202974759042263,
|
|
"num_tokens": 2029534.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.090177202224732,
|
|
"epoch": 0.9626128061882252,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998608654790741,
|
|
"loss": 6.7773,
|
|
"mean_token_accuracy": 0.11798964366316796,
|
|
"num_tokens": 2039143.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.939780521392822,
|
|
"epoch": 0.9669101847872797,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499848929208808,
|
|
"loss": 6.7368,
|
|
"mean_token_accuracy": 0.12413144260644912,
|
|
"num_tokens": 2048253.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.968171167373657,
|
|
"epoch": 0.9712075633863343,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998365019583519,
|
|
"loss": 6.7792,
|
|
"mean_token_accuracy": 0.12741474211215972,
|
|
"num_tokens": 2057234.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.055770301818848,
|
|
"epoch": 0.975504941985389,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998235837548417,
|
|
"loss": 6.8277,
|
|
"mean_token_accuracy": 0.12360714524984359,
|
|
"num_tokens": 2065431.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.023305130004883,
|
|
"epoch": 0.9798023205844435,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499810174626486,
|
|
"loss": 6.8226,
|
|
"mean_token_accuracy": 0.12461819276213645,
|
|
"num_tokens": 2074723.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.984486246109009,
|
|
"epoch": 0.984099699183498,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997962746025646,
|
|
"loss": 6.7019,
|
|
"mean_token_accuracy": 0.13492951169610023,
|
|
"num_tokens": 2084509.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.0700774669647215,
|
|
"epoch": 0.9883970777825526,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004997818837134298,
|
|
"loss": 6.849,
|
|
"mean_token_accuracy": 0.12504945918917656,
|
|
"num_tokens": 2093110.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.870955991744995,
|
|
"epoch": 0.9926944563816072,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997670019905057,
|
|
"loss": 6.702,
|
|
"mean_token_accuracy": 0.1213143840432167,
|
|
"num_tokens": 2102355.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.033004474639893,
|
|
"epoch": 0.9969918349806618,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997516294662876,
|
|
"loss": 6.7378,
|
|
"mean_token_accuracy": 0.12335852757096291,
|
|
"num_tokens": 2110418.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.9259838528103295,
|
|
"epoch": 1.0008594757198108,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997357661743433,
|
|
"loss": 6.717,
|
|
"mean_token_accuracy": 0.128744567433993,
|
|
"num_tokens": 2117866.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.995123481750488,
|
|
"epoch": 1.0051568543188656,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997194121493118,
|
|
"loss": 6.5734,
|
|
"mean_token_accuracy": 0.12787071764469146,
|
|
"num_tokens": 2126082.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.919755554199218,
|
|
"epoch": 1.0094542329179201,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997025674269037,
|
|
"loss": 6.5449,
|
|
"mean_token_accuracy": 0.1354580543935299,
|
|
"num_tokens": 2134042.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.861895275115967,
|
|
"epoch": 1.0137516115169747,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996852320439013,
|
|
"loss": 6.6139,
|
|
"mean_token_accuracy": 0.12945729792118071,
|
|
"num_tokens": 2142570.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.911503505706787,
|
|
"epoch": 1.0180489901160292,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996674060381578,
|
|
"loss": 6.5633,
|
|
"mean_token_accuracy": 0.13328466117382048,
|
|
"num_tokens": 2151310.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.907685422897339,
|
|
"epoch": 1.0223463687150838,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996490894485985,
|
|
"loss": 6.6286,
|
|
"mean_token_accuracy": 0.12506390661001204,
|
|
"num_tokens": 2160662.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.932943487167359,
|
|
"epoch": 1.0266437473141383,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996302823152193,
|
|
"loss": 6.566,
|
|
"mean_token_accuracy": 0.1292807199060917,
|
|
"num_tokens": 2170067.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.858935070037842,
|
|
"epoch": 1.0309411259131929,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996109846790873,
|
|
"loss": 6.5326,
|
|
"mean_token_accuracy": 0.1338893659412861,
|
|
"num_tokens": 2178850.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.8632307052612305,
|
|
"epoch": 1.0352385045122476,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004995911965823412,
|
|
"loss": 6.5605,
|
|
"mean_token_accuracy": 0.13701220229268074,
|
|
"num_tokens": 2188307.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.856139659881592,
|
|
"epoch": 1.0395358831113022,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995709180681899,
|
|
"loss": 6.5326,
|
|
"mean_token_accuracy": 0.140198864787817,
|
|
"num_tokens": 2197026.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.900337648391724,
|
|
"epoch": 1.0438332617103567,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499550149180914,
|
|
"loss": 6.5314,
|
|
"mean_token_accuracy": 0.13545763865113258,
|
|
"num_tokens": 2205537.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.889223337173462,
|
|
"epoch": 1.0481306403094113,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995288899658641,
|
|
"loss": 6.5543,
|
|
"mean_token_accuracy": 0.13490064293146134,
|
|
"num_tokens": 2214508.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.821092844009399,
|
|
"epoch": 1.0524280189084658,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995071404694619,
|
|
"loss": 6.6698,
|
|
"mean_token_accuracy": 0.1264920301735401,
|
|
"num_tokens": 2223084.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.998721551895142,
|
|
"epoch": 1.0567253975075204,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994849007391996,
|
|
"loss": 6.5995,
|
|
"mean_token_accuracy": 0.1271670438349247,
|
|
"num_tokens": 2231406.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.76812891960144,
|
|
"epoch": 1.061022776106575,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994621708236401,
|
|
"loss": 6.5119,
|
|
"mean_token_accuracy": 0.131594667583704,
|
|
"num_tokens": 2239867.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.892703628540039,
|
|
"epoch": 1.0653201547056295,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499438950772416,
|
|
"loss": 6.5595,
|
|
"mean_token_accuracy": 0.13020255863666536,
|
|
"num_tokens": 2248844.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.80609540939331,
|
|
"epoch": 1.0696175333046842,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004994152406362311,
|
|
"loss": 6.4868,
|
|
"mean_token_accuracy": 0.1369902640581131,
|
|
"num_tokens": 2257599.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.888293266296387,
|
|
"epoch": 1.0739149119037388,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004993910404668586,
|
|
"loss": 6.5419,
|
|
"mean_token_accuracy": 0.12804562970995903,
|
|
"num_tokens": 2266510.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.807245254516602,
|
|
"epoch": 1.0782122905027933,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499366350317142,
|
|
"loss": 6.5445,
|
|
"mean_token_accuracy": 0.13409808352589608,
|
|
"num_tokens": 2275462.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.824594402313233,
|
|
"epoch": 1.0825096691018479,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993411702409948,
|
|
"loss": 6.5053,
|
|
"mean_token_accuracy": 0.13172662556171416,
|
|
"num_tokens": 2283826.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.823530054092407,
|
|
"epoch": 1.0868070477009024,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993155002934002,
|
|
"loss": 6.5131,
|
|
"mean_token_accuracy": 0.1332196593284607,
|
|
"num_tokens": 2292967.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.906051445007324,
|
|
"epoch": 1.091104426299957,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004992893405304111,
|
|
"loss": 6.6504,
|
|
"mean_token_accuracy": 0.13221597969532012,
|
|
"num_tokens": 2302336.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.785560846328735,
|
|
"epoch": 1.0954018048990115,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00049926269100915,
|
|
"loss": 6.5522,
|
|
"mean_token_accuracy": 0.13710481822490692,
|
|
"num_tokens": 2311465.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.9224625587463375,
|
|
"epoch": 1.0996991834980663,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992355517878087,
|
|
"loss": 6.6554,
|
|
"mean_token_accuracy": 0.12584567815065384,
|
|
"num_tokens": 2320281.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.836811113357544,
|
|
"epoch": 1.1039965620971208,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004992079229256484,
|
|
"loss": 6.5584,
|
|
"mean_token_accuracy": 0.12910671755671502,
|
|
"num_tokens": 2329755.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.704578638076782,
|
|
"epoch": 1.1082939406961754,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004991798044829996,
|
|
"loss": 6.5024,
|
|
"mean_token_accuracy": 0.12955976650118828,
|
|
"num_tokens": 2338807.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.910696506500244,
|
|
"epoch": 1.11259131929523,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991511965212618,
|
|
"loss": 6.5914,
|
|
"mean_token_accuracy": 0.13309080600738527,
|
|
"num_tokens": 2348056.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.757680082321167,
|
|
"epoch": 1.1168886978942845,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004991220991029032,
|
|
"loss": 6.5917,
|
|
"mean_token_accuracy": 0.12701491191983222,
|
|
"num_tokens": 2357780.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.866291856765747,
|
|
"epoch": 1.121186076493339,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000499092512291461,
|
|
"loss": 6.5742,
|
|
"mean_token_accuracy": 0.1293620027601719,
|
|
"num_tokens": 2367060.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.836187791824341,
|
|
"epoch": 1.1254834550923936,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499062436151541,
|
|
"loss": 6.5736,
|
|
"mean_token_accuracy": 0.1331700198352337,
|
|
"num_tokens": 2375751.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.893719673156738,
|
|
"epoch": 1.129780833691448,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004990318707488173,
|
|
"loss": 6.6178,
|
|
"mean_token_accuracy": 0.1281808391213417,
|
|
"num_tokens": 2385013.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.797501945495606,
|
|
"epoch": 1.1340782122905029,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004990008161500327,
|
|
"loss": 6.5368,
|
|
"mean_token_accuracy": 0.1343330167233944,
|
|
"num_tokens": 2392935.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.829013967514038,
|
|
"epoch": 1.1383755908895574,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000498969272422998,
|
|
"loss": 6.532,
|
|
"mean_token_accuracy": 0.1358281135559082,
|
|
"num_tokens": 2401560.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.755909633636475,
|
|
"epoch": 1.142672969488612,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004989372396365921,
|
|
"loss": 6.4518,
|
|
"mean_token_accuracy": 0.13816773220896722,
|
|
"num_tokens": 2410050.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.890218067169189,
|
|
"epoch": 1.1469703480876665,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004989047178607618,
|
|
"loss": 6.547,
|
|
"mean_token_accuracy": 0.1365761749446392,
|
|
"num_tokens": 2418980.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.804721832275391,
|
|
"epoch": 1.151267726686721,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004988717071665215,
|
|
"loss": 6.5574,
|
|
"mean_token_accuracy": 0.1326664350926876,
|
|
"num_tokens": 2427992.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.721598196029663,
|
|
"epoch": 1.1555651052857756,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004988382076259537,
|
|
"loss": 6.4609,
|
|
"mean_token_accuracy": 0.138286504894495,
|
|
"num_tokens": 2436368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.781888771057129,
|
|
"epoch": 1.1598624838848304,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988042193122077,
|
|
"loss": 6.4618,
|
|
"mean_token_accuracy": 0.13687749654054643,
|
|
"num_tokens": 2445499.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.851740026473999,
|
|
"epoch": 1.164159862483885,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987697422995005,
|
|
"loss": 6.4922,
|
|
"mean_token_accuracy": 0.13173471242189408,
|
|
"num_tokens": 2454312.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.713117170333862,
|
|
"epoch": 1.1684572410829395,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004987347766631161,
|
|
"loss": 6.548,
|
|
"mean_token_accuracy": 0.1326726868748665,
|
|
"num_tokens": 2462922.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.791418552398682,
|
|
"epoch": 1.172754619681994,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004986993224794055,
|
|
"loss": 6.5917,
|
|
"mean_token_accuracy": 0.12623094841837884,
|
|
"num_tokens": 2472195.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.79559736251831,
|
|
"epoch": 1.1770519982810486,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004986633798257865,
|
|
"loss": 6.4975,
|
|
"mean_token_accuracy": 0.1346569150686264,
|
|
"num_tokens": 2481021.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.720452213287354,
|
|
"epoch": 1.181349376880103,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986269487807434,
|
|
"loss": 6.5027,
|
|
"mean_token_accuracy": 0.13570783659815788,
|
|
"num_tokens": 2490250.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.770303201675415,
|
|
"epoch": 1.1856467554791577,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000498590029423827,
|
|
"loss": 6.5757,
|
|
"mean_token_accuracy": 0.1362602196633816,
|
|
"num_tokens": 2499122.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.838917875289917,
|
|
"epoch": 1.1899441340782122,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985526218356546,
|
|
"loss": 6.5512,
|
|
"mean_token_accuracy": 0.1287482999265194,
|
|
"num_tokens": 2508454.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.7279455184936525,
|
|
"epoch": 1.1942415126772667,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004985147260979093,
|
|
"loss": 6.4843,
|
|
"mean_token_accuracy": 0.14064208790659904,
|
|
"num_tokens": 2517353.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.755608987808228,
|
|
"epoch": 1.1985388912763215,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004984763422933402,
|
|
"loss": 6.4984,
|
|
"mean_token_accuracy": 0.13640860766172408,
|
|
"num_tokens": 2526321.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.755157899856568,
|
|
"epoch": 1.202836269875376,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004984374705057623,
|
|
"loss": 6.5168,
|
|
"mean_token_accuracy": 0.13545396104454993,
|
|
"num_tokens": 2535924.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.818363904953003,
|
|
"epoch": 1.2071336484744306,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004983981108200561,
|
|
"loss": 6.5065,
|
|
"mean_token_accuracy": 0.1336505174636841,
|
|
"num_tokens": 2545606.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.692107009887695,
|
|
"epoch": 1.2114310270734852,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004983582633221672,
|
|
"loss": 6.5062,
|
|
"mean_token_accuracy": 0.1383553020656109,
|
|
"num_tokens": 2554947.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.83019061088562,
|
|
"epoch": 1.2157284056725397,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004983179280991068,
|
|
"loss": 6.6534,
|
|
"mean_token_accuracy": 0.12812817469239235,
|
|
"num_tokens": 2564462.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.796614646911621,
|
|
"epoch": 1.2200257842715942,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982771052389508,
|
|
"loss": 6.4877,
|
|
"mean_token_accuracy": 0.1379811257123947,
|
|
"num_tokens": 2573124.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.79015884399414,
|
|
"epoch": 1.224323162870649,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982357948308401,
|
|
"loss": 6.59,
|
|
"mean_token_accuracy": 0.12893555760383607,
|
|
"num_tokens": 2581829.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.829104900360107,
|
|
"epoch": 1.2286205414697036,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004981939969649799,
|
|
"loss": 6.4392,
|
|
"mean_token_accuracy": 0.13909578174352646,
|
|
"num_tokens": 2590631.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.743937397003174,
|
|
"epoch": 1.232917920068758,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004981517117326404,
|
|
"loss": 6.5519,
|
|
"mean_token_accuracy": 0.13750746101140976,
|
|
"num_tokens": 2600684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.769179630279541,
|
|
"epoch": 1.2372152986678127,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004981089392261553,
|
|
"loss": 6.4703,
|
|
"mean_token_accuracy": 0.14099505171179771,
|
|
"num_tokens": 2609667.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.69663987159729,
|
|
"epoch": 1.2415126772668672,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000498065679538923,
|
|
"loss": 6.5504,
|
|
"mean_token_accuracy": 0.13748323917388916,
|
|
"num_tokens": 2620025.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.7704236030578615,
|
|
"epoch": 1.2458100558659218,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004980219327654049,
|
|
"loss": 6.4502,
|
|
"mean_token_accuracy": 0.1409187763929367,
|
|
"num_tokens": 2629032.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.728482866287232,
|
|
"epoch": 1.2501074344649763,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000497977699001127,
|
|
"loss": 6.4341,
|
|
"mean_token_accuracy": 0.1414632149040699,
|
|
"num_tokens": 2638303.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.760922050476074,
|
|
"epoch": 1.2544048130640308,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004979329783426778,
|
|
"loss": 6.4682,
|
|
"mean_token_accuracy": 0.13974226340651513,
|
|
"num_tokens": 2647902.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.739649820327759,
|
|
"epoch": 1.2587021916630854,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004978877708877094,
|
|
"loss": 6.5202,
|
|
"mean_token_accuracy": 0.1345860943198204,
|
|
"num_tokens": 2657902.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.732415056228637,
|
|
"epoch": 1.2629995702621402,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004978420767349368,
|
|
"loss": 6.4395,
|
|
"mean_token_accuracy": 0.1351854085922241,
|
|
"num_tokens": 2667082.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.715655040740967,
|
|
"epoch": 1.2672969488611947,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004977958959841379,
|
|
"loss": 6.5146,
|
|
"mean_token_accuracy": 0.1359639957547188,
|
|
"num_tokens": 2676855.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.778389596939087,
|
|
"epoch": 1.2715943274602493,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497749228736153,
|
|
"loss": 6.4591,
|
|
"mean_token_accuracy": 0.13977851048111917,
|
|
"num_tokens": 2685750.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.708815574645996,
|
|
"epoch": 1.2758917060593038,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004977020750928845,
|
|
"loss": 6.5176,
|
|
"mean_token_accuracy": 0.14020123705267906,
|
|
"num_tokens": 2695272.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.7224524974822994,
|
|
"epoch": 1.2801890846583583,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004976544351572973,
|
|
"loss": 6.4449,
|
|
"mean_token_accuracy": 0.14301179200410843,
|
|
"num_tokens": 2704806.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.649525260925293,
|
|
"epoch": 1.2844864632574131,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004976063090334179,
|
|
"loss": 6.4976,
|
|
"mean_token_accuracy": 0.13876865953207015,
|
|
"num_tokens": 2713521.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.740707063674927,
|
|
"epoch": 1.2887838418564677,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004975576968263346,
|
|
"loss": 6.502,
|
|
"mean_token_accuracy": 0.12958571687340736,
|
|
"num_tokens": 2721848.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2887838418564677,
|
|
"eval_entropy": 6.470501623712145,
|
|
"eval_loss": 6.560330390930176,
|
|
"eval_mean_token_accuracy": 0.13803270239349422,
|
|
"eval_num_tokens": 2721848.0,
|
|
"eval_runtime": 2.0451,
|
|
"eval_samples_per_second": 1735.35,
|
|
"eval_steps_per_second": 217.102,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.670271635055542,
|
|
"epoch": 1.2930812204555222,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000497508598642197,
|
|
"loss": 6.4776,
|
|
"mean_token_accuracy": 0.1379353679716587,
|
|
"num_tokens": 2731473.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.75262417793274,
|
|
"epoch": 1.2973785990545768,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497459014588216,
|
|
"loss": 6.5328,
|
|
"mean_token_accuracy": 0.13872059807181358,
|
|
"num_tokens": 2739867.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.689033460617066,
|
|
"epoch": 1.3016759776536313,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497408944772663,
|
|
"loss": 6.4493,
|
|
"mean_token_accuracy": 0.13728251829743385,
|
|
"num_tokens": 2748903.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.651050281524658,
|
|
"epoch": 1.3059733562526858,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004973583893048707,
|
|
"loss": 6.4194,
|
|
"mean_token_accuracy": 0.1359269514679909,
|
|
"num_tokens": 2757711.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.832587623596192,
|
|
"epoch": 1.3102707348517404,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004973073482952321,
|
|
"loss": 6.4476,
|
|
"mean_token_accuracy": 0.1389494240283966,
|
|
"num_tokens": 2765633.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.610323524475097,
|
|
"epoch": 1.314568113450795,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004972558218552004,
|
|
"loss": 6.4865,
|
|
"mean_token_accuracy": 0.1345442496240139,
|
|
"num_tokens": 2774495.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.781571102142334,
|
|
"epoch": 1.3188654920498495,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004972038100972885,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.13351112902164458,
|
|
"num_tokens": 2782665.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.640726566314697,
|
|
"epoch": 1.323162870648904,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004971513131350697,
|
|
"loss": 6.4315,
|
|
"mean_token_accuracy": 0.13331406638026239,
|
|
"num_tokens": 2791394.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.637244319915771,
|
|
"epoch": 1.3274602492479588,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004970983310831759,
|
|
"loss": 6.4216,
|
|
"mean_token_accuracy": 0.13528204038739206,
|
|
"num_tokens": 2800488.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.73633337020874,
|
|
"epoch": 1.3317576278470133,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004970448640572989,
|
|
"loss": 6.554,
|
|
"mean_token_accuracy": 0.13508356660604476,
|
|
"num_tokens": 2810116.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.68259482383728,
|
|
"epoch": 1.336055006446068,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004969909121741895,
|
|
"loss": 6.362,
|
|
"mean_token_accuracy": 0.14365303963422776,
|
|
"num_tokens": 2819205.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.660860872268676,
|
|
"epoch": 1.3403523850451224,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004969364755516569,
|
|
"loss": 6.4237,
|
|
"mean_token_accuracy": 0.13722610473632812,
|
|
"num_tokens": 2828017.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.719718647003174,
|
|
"epoch": 1.344649763644177,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004968815543085689,
|
|
"loss": 6.4716,
|
|
"mean_token_accuracy": 0.1409387357532978,
|
|
"num_tokens": 2837125.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.677382087707519,
|
|
"epoch": 1.3489471422432318,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004968261485648516,
|
|
"loss": 6.4702,
|
|
"mean_token_accuracy": 0.1383634626865387,
|
|
"num_tokens": 2845438.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.7003734588623045,
|
|
"epoch": 1.3532445208422863,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000496770258441489,
|
|
"loss": 6.4578,
|
|
"mean_token_accuracy": 0.14429745227098464,
|
|
"num_tokens": 2854389.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.630741024017334,
|
|
"epoch": 1.3575418994413408,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004967138840605228,
|
|
"loss": 6.4274,
|
|
"mean_token_accuracy": 0.14213711023330688,
|
|
"num_tokens": 2863654.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.580131101608276,
|
|
"epoch": 1.3618392780403954,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000496657025545052,
|
|
"loss": 6.3333,
|
|
"mean_token_accuracy": 0.14458294957876205,
|
|
"num_tokens": 2872871.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.652856779098511,
|
|
"epoch": 1.36613665663945,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000496599683019233,
|
|
"loss": 6.4385,
|
|
"mean_token_accuracy": 0.1378958448767662,
|
|
"num_tokens": 2881140.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.743864488601685,
|
|
"epoch": 1.3704340352385045,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000496541856608279,
|
|
"loss": 6.4138,
|
|
"mean_token_accuracy": 0.13684486821293831,
|
|
"num_tokens": 2889945.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.543468856811524,
|
|
"epoch": 1.374731413837559,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004964835464384595,
|
|
"loss": 6.3442,
|
|
"mean_token_accuracy": 0.14644223153591157,
|
|
"num_tokens": 2898897.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.6516578674316404,
|
|
"epoch": 1.3790287924366136,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000496424752637101,
|
|
"loss": 6.3765,
|
|
"mean_token_accuracy": 0.14313278645277022,
|
|
"num_tokens": 2907717.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.6023597240448,
|
|
"epoch": 1.3833261710356681,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004963654753325853,
|
|
"loss": 6.3608,
|
|
"mean_token_accuracy": 0.142365200817585,
|
|
"num_tokens": 2916213.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.675694608688355,
|
|
"epoch": 1.387623549634723,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004963057146543505,
|
|
"loss": 6.5232,
|
|
"mean_token_accuracy": 0.13410560339689254,
|
|
"num_tokens": 2925706.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.640337657928467,
|
|
"epoch": 1.3919209282337774,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00049624547073289,
|
|
"loss": 6.4341,
|
|
"mean_token_accuracy": 0.1372315354645252,
|
|
"num_tokens": 2934464.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.644095182418823,
|
|
"epoch": 1.396218306832832,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004961847436997526,
|
|
"loss": 6.3392,
|
|
"mean_token_accuracy": 0.1437459670007229,
|
|
"num_tokens": 2944095.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.544746780395508,
|
|
"epoch": 1.4005156854318865,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004961235336875416,
|
|
"loss": 6.3337,
|
|
"mean_token_accuracy": 0.14818646609783173,
|
|
"num_tokens": 2953357.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.590663099288941,
|
|
"epoch": 1.404813064030941,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004960618408299154,
|
|
"loss": 6.4913,
|
|
"mean_token_accuracy": 0.13292545676231385,
|
|
"num_tokens": 2963020.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.676125764846802,
|
|
"epoch": 1.4091104426299956,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004959996652615865,
|
|
"loss": 6.3291,
|
|
"mean_token_accuracy": 0.1426067478954792,
|
|
"num_tokens": 2971955.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.682071399688721,
|
|
"epoch": 1.4134078212290504,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004959370071183216,
|
|
"loss": 6.3941,
|
|
"mean_token_accuracy": 0.1417133815586567,
|
|
"num_tokens": 2980662.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.630775785446167,
|
|
"epoch": 1.417705199828105,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004958738665369407,
|
|
"loss": 6.5253,
|
|
"mean_token_accuracy": 0.12776071205735207,
|
|
"num_tokens": 2990038.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.679714679718018,
|
|
"epoch": 1.4220025784271595,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004958102436553179,
|
|
"loss": 6.4334,
|
|
"mean_token_accuracy": 0.1367899753153324,
|
|
"num_tokens": 2999835.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.633715724945068,
|
|
"epoch": 1.426299957026214,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00049574613861238,
|
|
"loss": 6.428,
|
|
"mean_token_accuracy": 0.13732998967170715,
|
|
"num_tokens": 3009593.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.719851779937744,
|
|
"epoch": 1.4305973356252686,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004956815515481069,
|
|
"loss": 6.4586,
|
|
"mean_token_accuracy": 0.14100154265761375,
|
|
"num_tokens": 3019187.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.548164558410645,
|
|
"epoch": 1.4348947142243231,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004956164826035309,
|
|
"loss": 6.3723,
|
|
"mean_token_accuracy": 0.1403822347521782,
|
|
"num_tokens": 3027875.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.590739965438843,
|
|
"epoch": 1.4391920928233777,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004955509319207363,
|
|
"loss": 6.3917,
|
|
"mean_token_accuracy": 0.139241461455822,
|
|
"num_tokens": 3036902.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.564586973190307,
|
|
"epoch": 1.4434894714224322,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004954848996428601,
|
|
"loss": 6.37,
|
|
"mean_token_accuracy": 0.1485365867614746,
|
|
"num_tokens": 3046653.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.688973379135132,
|
|
"epoch": 1.4477868500214868,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00049541838591409,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.13771305233240128,
|
|
"num_tokens": 3056273.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.601505422592163,
|
|
"epoch": 1.4520842286205415,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004953513908796657,
|
|
"loss": 6.3754,
|
|
"mean_token_accuracy": 0.142202128469944,
|
|
"num_tokens": 3065662.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.688344573974609,
|
|
"epoch": 1.456381607219596,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004952839146858773,
|
|
"loss": 6.4125,
|
|
"mean_token_accuracy": 0.14289402961730957,
|
|
"num_tokens": 3073970.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.557283878326416,
|
|
"epoch": 1.4606789858186506,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004952159574800658,
|
|
"loss": 6.4149,
|
|
"mean_token_accuracy": 0.1368069365620613,
|
|
"num_tokens": 3082500.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.6616050720214846,
|
|
"epoch": 1.4649763644177052,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004951475194106229,
|
|
"loss": 6.3624,
|
|
"mean_token_accuracy": 0.14598480761051177,
|
|
"num_tokens": 3091574.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.564044332504272,
|
|
"epoch": 1.4692737430167597,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004950786006269898,
|
|
"loss": 6.4712,
|
|
"mean_token_accuracy": 0.13445298001170158,
|
|
"num_tokens": 3102402.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.630729532241821,
|
|
"epoch": 1.4735711216158143,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004950092012796576,
|
|
"loss": 6.2883,
|
|
"mean_token_accuracy": 0.14547964110970496,
|
|
"num_tokens": 3111347.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.57949857711792,
|
|
"epoch": 1.477868500214869,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004949393215201666,
|
|
"loss": 6.3576,
|
|
"mean_token_accuracy": 0.14195869863033295,
|
|
"num_tokens": 3120018.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.621220111846924,
|
|
"epoch": 1.4821658788139236,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004948689615011065,
|
|
"loss": 6.4303,
|
|
"mean_token_accuracy": 0.13502154499292374,
|
|
"num_tokens": 3129669.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.622284650802612,
|
|
"epoch": 1.4864632574129781,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004947981213761154,
|
|
"loss": 6.3626,
|
|
"mean_token_accuracy": 0.14655405953526496,
|
|
"num_tokens": 3139112.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.566265106201172,
|
|
"epoch": 1.4907606360120327,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004947268012998797,
|
|
"loss": 6.314,
|
|
"mean_token_accuracy": 0.15282158702611923,
|
|
"num_tokens": 3148437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.555867433547974,
|
|
"epoch": 1.4950580146110872,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000494655001428134,
|
|
"loss": 6.3163,
|
|
"mean_token_accuracy": 0.14129877462983131,
|
|
"num_tokens": 3158165.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.610077142715454,
|
|
"epoch": 1.4993553932101418,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004945827219176604,
|
|
"loss": 6.3806,
|
|
"mean_token_accuracy": 0.1494217187166214,
|
|
"num_tokens": 3167262.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.497924327850342,
|
|
"epoch": 1.5036527718091963,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004945099629262888,
|
|
"loss": 6.365,
|
|
"mean_token_accuracy": 0.14257355630397797,
|
|
"num_tokens": 3176696.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.763267278671265,
|
|
"epoch": 1.5079501504082509,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004944367246128954,
|
|
"loss": 6.4414,
|
|
"mean_token_accuracy": 0.13902869820594788,
|
|
"num_tokens": 3185857.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.499178218841553,
|
|
"epoch": 1.5122475290073054,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004943630071374036,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.15046660900115966,
|
|
"num_tokens": 3194687.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.597637414932251,
|
|
"epoch": 1.51654490760636,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004942888106607828,
|
|
"loss": 6.3408,
|
|
"mean_token_accuracy": 0.14210622310638427,
|
|
"num_tokens": 3204913.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.629096126556396,
|
|
"epoch": 1.5208422862054147,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004942141353450486,
|
|
"loss": 6.3319,
|
|
"mean_token_accuracy": 0.14792249947786332,
|
|
"num_tokens": 3213312.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.488542318344116,
|
|
"epoch": 1.5251396648044693,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004941389813532619,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.15323352664709092,
|
|
"num_tokens": 3222992.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.508915090560913,
|
|
"epoch": 1.5294370434035238,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000494063348849529,
|
|
"loss": 6.3006,
|
|
"mean_token_accuracy": 0.15017192512750627,
|
|
"num_tokens": 3232836.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.618602657318116,
|
|
"epoch": 1.5337344220025786,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004939872379990011,
|
|
"loss": 6.4568,
|
|
"mean_token_accuracy": 0.1389564484357834,
|
|
"num_tokens": 3243171.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.68247127532959,
|
|
"epoch": 1.5380318006016331,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004939106489678739,
|
|
"loss": 6.3675,
|
|
"mean_token_accuracy": 0.1462523892521858,
|
|
"num_tokens": 3251995.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.5040504932403564,
|
|
"epoch": 1.5423291792006877,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000493833581923387,
|
|
"loss": 6.3162,
|
|
"mean_token_accuracy": 0.14888131469488144,
|
|
"num_tokens": 3260841.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.610679483413696,
|
|
"epoch": 1.5466265577997422,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004937560370338244,
|
|
"loss": 6.4592,
|
|
"mean_token_accuracy": 0.13561476543545722,
|
|
"num_tokens": 3270979.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.649268531799317,
|
|
"epoch": 1.5509239363987968,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000493678014468513,
|
|
"loss": 6.3782,
|
|
"mean_token_accuracy": 0.15145477950572966,
|
|
"num_tokens": 3279848.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.525541353225708,
|
|
"epoch": 1.5552213149978513,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004935995143978227,
|
|
"loss": 6.3828,
|
|
"mean_token_accuracy": 0.14420599341392518,
|
|
"num_tokens": 3289172.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.548158836364746,
|
|
"epoch": 1.5595186935969059,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004935205369931664,
|
|
"loss": 6.2838,
|
|
"mean_token_accuracy": 0.1488450989127159,
|
|
"num_tokens": 3297432.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.629633808135987,
|
|
"epoch": 1.5638160721959604,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004934410824269992,
|
|
"loss": 6.307,
|
|
"mean_token_accuracy": 0.1468965768814087,
|
|
"num_tokens": 3307486.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.505390453338623,
|
|
"epoch": 1.568113450795015,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004933611508728182,
|
|
"loss": 6.277,
|
|
"mean_token_accuracy": 0.15054729878902434,
|
|
"num_tokens": 3316296.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.567506504058838,
|
|
"epoch": 1.5724108293940695,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000493280742505162,
|
|
"loss": 6.3093,
|
|
"mean_token_accuracy": 0.1449831858277321,
|
|
"num_tokens": 3326080.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.486201763153076,
|
|
"epoch": 1.576708207993124,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004931998574996102,
|
|
"loss": 6.2364,
|
|
"mean_token_accuracy": 0.14821842014789582,
|
|
"num_tokens": 3334826.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.458436775207519,
|
|
"epoch": 1.5810055865921788,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004931184960327832,
|
|
"loss": 6.243,
|
|
"mean_token_accuracy": 0.15286262780427934,
|
|
"num_tokens": 3343261.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.5017963409423825,
|
|
"epoch": 1.5853029651912334,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0004930366582823421,
|
|
"loss": 6.2846,
|
|
"mean_token_accuracy": 0.1460932895541191,
|
|
"num_tokens": 3352513.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.565137147903442,
|
|
"epoch": 1.589600343790288,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004929543444269879,
|
|
"loss": 6.3357,
|
|
"mean_token_accuracy": 0.14929510578513144,
|
|
"num_tokens": 3361577.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.485863208770752,
|
|
"epoch": 1.5938977223893425,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492871554646461,
|
|
"loss": 6.3948,
|
|
"mean_token_accuracy": 0.14030372574925423,
|
|
"num_tokens": 3370591.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.554273700714111,
|
|
"epoch": 1.5981951009883972,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004927882891215413,
|
|
"loss": 6.306,
|
|
"mean_token_accuracy": 0.14827967956662177,
|
|
"num_tokens": 3379761.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.592674922943115,
|
|
"epoch": 1.6024924795874518,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004927045480340475,
|
|
"loss": 6.3969,
|
|
"mean_token_accuracy": 0.1377456396818161,
|
|
"num_tokens": 3388974.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.534068393707275,
|
|
"epoch": 1.6067898581865063,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004926203315668363,
|
|
"loss": 6.3089,
|
|
"mean_token_accuracy": 0.14855122640728952,
|
|
"num_tokens": 3398339.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.51072096824646,
|
|
"epoch": 1.6110872367855609,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004925356399038032,
|
|
"loss": 6.2784,
|
|
"mean_token_accuracy": 0.1444276086986065,
|
|
"num_tokens": 3408292.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.546713066101074,
|
|
"epoch": 1.6153846153846154,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004924504732298808,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.15252432748675346,
|
|
"num_tokens": 3417057.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.566928911209106,
|
|
"epoch": 1.61968199398367,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004923648317310391,
|
|
"loss": 6.3645,
|
|
"mean_token_accuracy": 0.14241925179958342,
|
|
"num_tokens": 3425830.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.581201171875,
|
|
"epoch": 1.6239793725827245,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004922787155942849,
|
|
"loss": 6.3913,
|
|
"mean_token_accuracy": 0.13747936934232713,
|
|
"num_tokens": 3435513.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.493123197555542,
|
|
"epoch": 1.628276751181779,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004921921250076611,
|
|
"loss": 6.3081,
|
|
"mean_token_accuracy": 0.1506146050989628,
|
|
"num_tokens": 3444684.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.519382381439209,
|
|
"epoch": 1.6325741297808336,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004921050601602475,
|
|
"loss": 6.358,
|
|
"mean_token_accuracy": 0.14751976430416108,
|
|
"num_tokens": 3453454.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.561795377731324,
|
|
"epoch": 1.6368715083798882,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004920175212421587,
|
|
"loss": 6.2876,
|
|
"mean_token_accuracy": 0.1480035036802292,
|
|
"num_tokens": 3463228.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.427498245239258,
|
|
"epoch": 1.6411688869789427,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004919295084445445,
|
|
"loss": 6.1795,
|
|
"mean_token_accuracy": 0.15354781299829484,
|
|
"num_tokens": 3472131.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.453372526168823,
|
|
"epoch": 1.6454662655779975,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004918410219595899,
|
|
"loss": 6.262,
|
|
"mean_token_accuracy": 0.1537986733019352,
|
|
"num_tokens": 3480642.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.673685932159424,
|
|
"epoch": 1.649763644177052,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000491752061980514,
|
|
"loss": 6.2464,
|
|
"mean_token_accuracy": 0.15058573931455613,
|
|
"num_tokens": 3489346.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.364662742614746,
|
|
"epoch": 1.6540610227761066,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004916626287015697,
|
|
"loss": 6.2797,
|
|
"mean_token_accuracy": 0.14565936028957366,
|
|
"num_tokens": 3498473.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.596463584899903,
|
|
"epoch": 1.658358401375161,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004915727223180436,
|
|
"loss": 6.2889,
|
|
"mean_token_accuracy": 0.14369736313819886,
|
|
"num_tokens": 3507415.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.553056049346924,
|
|
"epoch": 1.6626557799742159,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004914823430262554,
|
|
"loss": 6.3913,
|
|
"mean_token_accuracy": 0.13537019938230516,
|
|
"num_tokens": 3516873.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.495272874832153,
|
|
"epoch": 1.6669531585732704,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004913914910235573,
|
|
"loss": 6.2558,
|
|
"mean_token_accuracy": 0.1485748916864395,
|
|
"num_tokens": 3525047.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.465020227432251,
|
|
"epoch": 1.671250537172325,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004913001665083337,
|
|
"loss": 6.2614,
|
|
"mean_token_accuracy": 0.1493809461593628,
|
|
"num_tokens": 3534354.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.480109977722168,
|
|
"epoch": 1.6755479157713795,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004912083696800008,
|
|
"loss": 6.3061,
|
|
"mean_token_accuracy": 0.14220771193504333,
|
|
"num_tokens": 3543830.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.481950473785401,
|
|
"epoch": 1.679845294370434,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004911161007390063,
|
|
"loss": 6.214,
|
|
"mean_token_accuracy": 0.1464408040046692,
|
|
"num_tokens": 3552314.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.433442115783691,
|
|
"epoch": 1.6841426729694886,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004910233598868287,
|
|
"loss": 6.2733,
|
|
"mean_token_accuracy": 0.14608966633677484,
|
|
"num_tokens": 3561656.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.5319255828857425,
|
|
"epoch": 1.6884400515685432,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004909301473259769,
|
|
"loss": 6.2782,
|
|
"mean_token_accuracy": 0.14630958884954454,
|
|
"num_tokens": 3571784.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.498413991928101,
|
|
"epoch": 1.6927374301675977,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004908364632599899,
|
|
"loss": 6.2466,
|
|
"mean_token_accuracy": 0.14925297126173973,
|
|
"num_tokens": 3580626.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.409846687316895,
|
|
"epoch": 1.6970348087666522,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004907423078934362,
|
|
"loss": 6.2432,
|
|
"mean_token_accuracy": 0.14702003747224807,
|
|
"num_tokens": 3589916.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.491791534423828,
|
|
"epoch": 1.7013321873657068,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004906476814319134,
|
|
"loss": 6.2665,
|
|
"mean_token_accuracy": 0.1481113463640213,
|
|
"num_tokens": 3599128.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.418525266647339,
|
|
"epoch": 1.7056295659647613,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004905525840820481,
|
|
"loss": 6.2822,
|
|
"mean_token_accuracy": 0.14321094751358032,
|
|
"num_tokens": 3608764.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.584860706329346,
|
|
"epoch": 1.709926944563816,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004904570160514948,
|
|
"loss": 6.318,
|
|
"mean_token_accuracy": 0.14112781435251237,
|
|
"num_tokens": 3619082.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.488425922393799,
|
|
"epoch": 1.7142243231628707,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004903609775489358,
|
|
"loss": 6.2898,
|
|
"mean_token_accuracy": 0.1483103021979332,
|
|
"num_tokens": 3628695.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.527116823196411,
|
|
"epoch": 1.7185217017619252,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004902644687840809,
|
|
"loss": 6.2518,
|
|
"mean_token_accuracy": 0.14549941420555115,
|
|
"num_tokens": 3637599.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7185217017619252,
|
|
"eval_entropy": 6.19347925014324,
|
|
"eval_loss": 6.329723358154297,
|
|
"eval_mean_token_accuracy": 0.1496910023091881,
|
|
"eval_num_tokens": 3637599.0,
|
|
"eval_runtime": 2.05,
|
|
"eval_samples_per_second": 1731.192,
|
|
"eval_steps_per_second": 216.582,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.444890356063842,
|
|
"epoch": 1.7228190803609797,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004901674899676667,
|
|
"loss": 6.2616,
|
|
"mean_token_accuracy": 0.1490693584084511,
|
|
"num_tokens": 3647406.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.4421497821807865,
|
|
"epoch": 1.7271164589600345,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004900700413114561,
|
|
"loss": 6.1411,
|
|
"mean_token_accuracy": 0.1515228644013405,
|
|
"num_tokens": 3656531.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.438943338394165,
|
|
"epoch": 1.731413837559089,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000489972123028238,
|
|
"loss": 6.2278,
|
|
"mean_token_accuracy": 0.1497970812022686,
|
|
"num_tokens": 3664922.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.477704477310181,
|
|
"epoch": 1.7357112161581436,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004898737353318268,
|
|
"loss": 6.1731,
|
|
"mean_token_accuracy": 0.1531771868467331,
|
|
"num_tokens": 3673283.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.417043018341064,
|
|
"epoch": 1.7400085947571982,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489774878437062,
|
|
"loss": 6.3189,
|
|
"mean_token_accuracy": 0.1491183653473854,
|
|
"num_tokens": 3681760.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.496932172775269,
|
|
"epoch": 1.7443059733562527,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004896755525598074,
|
|
"loss": 6.137,
|
|
"mean_token_accuracy": 0.15091845840215684,
|
|
"num_tokens": 3689408.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.424662256240845,
|
|
"epoch": 1.7486033519553073,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895757579169511,
|
|
"loss": 6.2482,
|
|
"mean_token_accuracy": 0.14746748507022858,
|
|
"num_tokens": 3697904.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.576484441757202,
|
|
"epoch": 1.7529007305543618,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004894754947264047,
|
|
"loss": 6.2665,
|
|
"mean_token_accuracy": 0.14906733110547066,
|
|
"num_tokens": 3706704.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.423267459869384,
|
|
"epoch": 1.7571981091534163,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000489374763207103,
|
|
"loss": 6.3263,
|
|
"mean_token_accuracy": 0.14524614065885544,
|
|
"num_tokens": 3715690.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.511725521087646,
|
|
"epoch": 1.761495487752471,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004892735635790033,
|
|
"loss": 6.1332,
|
|
"mean_token_accuracy": 0.1547706514596939,
|
|
"num_tokens": 3724835.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.357218599319458,
|
|
"epoch": 1.7657928663515254,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000489171896063085,
|
|
"loss": 6.1786,
|
|
"mean_token_accuracy": 0.15807082131505013,
|
|
"num_tokens": 3733977.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.503004789352417,
|
|
"epoch": 1.77009024495058,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004890697608813495,
|
|
"loss": 6.2892,
|
|
"mean_token_accuracy": 0.14620151743292809,
|
|
"num_tokens": 3742665.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.575605297088623,
|
|
"epoch": 1.7743876235496348,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004889671582568193,
|
|
"loss": 6.3394,
|
|
"mean_token_accuracy": 0.14444702491164207,
|
|
"num_tokens": 3751647.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.386636257171631,
|
|
"epoch": 1.7786850021486893,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004888640884135374,
|
|
"loss": 6.2572,
|
|
"mean_token_accuracy": 0.14602783024311067,
|
|
"num_tokens": 3760852.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.443797779083252,
|
|
"epoch": 1.7829823807477438,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004887605515765671,
|
|
"loss": 6.1999,
|
|
"mean_token_accuracy": 0.1500324159860611,
|
|
"num_tokens": 3768640.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.453762483596802,
|
|
"epoch": 1.7872797593467986,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004886565479719914,
|
|
"loss": 6.2371,
|
|
"mean_token_accuracy": 0.14520688354969025,
|
|
"num_tokens": 3776504.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.546750164031982,
|
|
"epoch": 1.7915771379458532,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004885520778269128,
|
|
"loss": 6.2633,
|
|
"mean_token_accuracy": 0.1508520394563675,
|
|
"num_tokens": 3786353.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.421859884262085,
|
|
"epoch": 1.7958745165449077,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004884471413694523,
|
|
"loss": 6.2927,
|
|
"mean_token_accuracy": 0.14654150381684303,
|
|
"num_tokens": 3795902.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.475854921340942,
|
|
"epoch": 1.8001718951439623,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004883417388287491,
|
|
"loss": 6.2152,
|
|
"mean_token_accuracy": 0.14346302896738053,
|
|
"num_tokens": 3805986.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.390176630020141,
|
|
"epoch": 1.8044692737430168,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004882358704349603,
|
|
"loss": 6.3309,
|
|
"mean_token_accuracy": 0.15342760533094407,
|
|
"num_tokens": 3814915.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.5035735130310055,
|
|
"epoch": 1.8087666523420713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004881295364192601,
|
|
"loss": 6.2208,
|
|
"mean_token_accuracy": 0.15399999171495438,
|
|
"num_tokens": 3823966.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.512830305099487,
|
|
"epoch": 1.813064030941126,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004880227370138394,
|
|
"loss": 6.2733,
|
|
"mean_token_accuracy": 0.141741144657135,
|
|
"num_tokens": 3832775.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.412053823471069,
|
|
"epoch": 1.8173614095401804,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004879154724519057,
|
|
"loss": 6.1883,
|
|
"mean_token_accuracy": 0.15187056958675385,
|
|
"num_tokens": 3842808.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.525836372375489,
|
|
"epoch": 1.821658788139235,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004878077429676816,
|
|
"loss": 6.3298,
|
|
"mean_token_accuracy": 0.14646431654691697,
|
|
"num_tokens": 3853303.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.404772138595581,
|
|
"epoch": 1.8259561667382895,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004876995487964054,
|
|
"loss": 6.2232,
|
|
"mean_token_accuracy": 0.14396868348121644,
|
|
"num_tokens": 3862462.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.504109525680542,
|
|
"epoch": 1.830253545337344,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048759089017432996,
|
|
"loss": 6.3474,
|
|
"mean_token_accuracy": 0.1436037801206112,
|
|
"num_tokens": 3871596.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.516725015640259,
|
|
"epoch": 1.8345509239363988,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004874817673387222,
|
|
"loss": 6.2488,
|
|
"mean_token_accuracy": 0.150234591960907,
|
|
"num_tokens": 3881276.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.40357723236084,
|
|
"epoch": 1.8388483025354534,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048737218052786275,
|
|
"loss": 6.3306,
|
|
"mean_token_accuracy": 0.1420318104326725,
|
|
"num_tokens": 3891610.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.566394519805908,
|
|
"epoch": 1.843145681134508,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048726212998104554,
|
|
"loss": 6.2537,
|
|
"mean_token_accuracy": 0.14613947346806527,
|
|
"num_tokens": 3900584.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.378025674819947,
|
|
"epoch": 1.8474430597335625,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004871516159385768,
|
|
"loss": 6.2048,
|
|
"mean_token_accuracy": 0.15661944225430488,
|
|
"num_tokens": 3910208.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.331248712539673,
|
|
"epoch": 1.8517404383326173,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004870406386417752,
|
|
"loss": 6.1133,
|
|
"mean_token_accuracy": 0.15902349054813386,
|
|
"num_tokens": 3918424.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.3689392566680905,
|
|
"epoch": 1.8560378169316718,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004869291983329707,
|
|
"loss": 6.0464,
|
|
"mean_token_accuracy": 0.16715734601020812,
|
|
"num_tokens": 3926206.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.45319676399231,
|
|
"epoch": 1.8603351955307263,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004868172952555044,
|
|
"loss": 6.1646,
|
|
"mean_token_accuracy": 0.14661258310079575,
|
|
"num_tokens": 3935769.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.394322872161865,
|
|
"epoch": 1.864632574129781,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004867049296537278,
|
|
"loss": 6.1415,
|
|
"mean_token_accuracy": 0.15269692093133927,
|
|
"num_tokens": 3945118.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.474433708190918,
|
|
"epoch": 1.8689299527288354,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004865921017730027,
|
|
"loss": 6.2316,
|
|
"mean_token_accuracy": 0.15484376698732377,
|
|
"num_tokens": 3954012.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.439454174041748,
|
|
"epoch": 1.87322733132789,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048647881185969995,
|
|
"loss": 6.2453,
|
|
"mean_token_accuracy": 0.14685868918895723,
|
|
"num_tokens": 3964239.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.393148756027221,
|
|
"epoch": 1.8775247099269445,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004863650601611994,
|
|
"loss": 6.1549,
|
|
"mean_token_accuracy": 0.15369305908679962,
|
|
"num_tokens": 3973694.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.382004022598267,
|
|
"epoch": 1.881822088525999,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048625084692588937,
|
|
"loss": 6.1903,
|
|
"mean_token_accuracy": 0.15866641253232955,
|
|
"num_tokens": 3982706.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.365177822113037,
|
|
"epoch": 1.8861194671250536,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048613617240316593,
|
|
"loss": 6.1493,
|
|
"mean_token_accuracy": 0.15633954405784606,
|
|
"num_tokens": 3990934.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.4114241123199465,
|
|
"epoch": 1.8904168457241082,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004860210368434323,
|
|
"loss": 6.202,
|
|
"mean_token_accuracy": 0.15539102777838706,
|
|
"num_tokens": 3999864.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.44605655670166,
|
|
"epoch": 1.8947142243231627,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00048590544049809857,
|
|
"loss": 6.2143,
|
|
"mean_token_accuracy": 0.15502872616052626,
|
|
"num_tokens": 4008273.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.396330881118774,
|
|
"epoch": 1.8990116029222175,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000485789383619581,
|
|
"loss": 6.2196,
|
|
"mean_token_accuracy": 0.1552313044667244,
|
|
"num_tokens": 4017697.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.393840074539185,
|
|
"epoch": 1.903308981521272,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004856728664613015,
|
|
"loss": 6.2381,
|
|
"mean_token_accuracy": 0.1517387181520462,
|
|
"num_tokens": 4026775.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.4121527671813965,
|
|
"epoch": 1.9076063601203266,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048555588927768674,
|
|
"loss": 6.2065,
|
|
"mean_token_accuracy": 0.15306443050503732,
|
|
"num_tokens": 4036476.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.4631853103637695,
|
|
"epoch": 1.9119037387193811,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004854384523241683,
|
|
"loss": 6.1853,
|
|
"mean_token_accuracy": 0.14910603538155556,
|
|
"num_tokens": 4045221.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.358554410934448,
|
|
"epoch": 1.916201117318436,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048532055585718143,
|
|
"loss": 6.1143,
|
|
"mean_token_accuracy": 0.1558242991566658,
|
|
"num_tokens": 4053754.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.419343090057373,
|
|
"epoch": 1.9204984959174904,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048520220013416505,
|
|
"loss": 6.1398,
|
|
"mean_token_accuracy": 0.15855558216571808,
|
|
"num_tokens": 4061730.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.371563529968261,
|
|
"epoch": 1.924795874516545,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004850833854135607,
|
|
"loss": 6.2178,
|
|
"mean_token_accuracy": 0.15214873254299163,
|
|
"num_tokens": 4070501.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.432424831390381,
|
|
"epoch": 1.9290932531155995,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004849641119548122,
|
|
"loss": 6.284,
|
|
"mean_token_accuracy": 0.14980236291885377,
|
|
"num_tokens": 4079621.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.53593487739563,
|
|
"epoch": 1.933390631714654,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000484844380018365,
|
|
"loss": 6.2792,
|
|
"mean_token_accuracy": 0.14946389943361282,
|
|
"num_tokens": 4090106.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.445489692687988,
|
|
"epoch": 1.9376880103137086,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000484724189865666,
|
|
"loss": 6.2137,
|
|
"mean_token_accuracy": 0.15263792276382446,
|
|
"num_tokens": 4099269.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.2474573135375975,
|
|
"epoch": 1.9419853889127632,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004846035417591624,
|
|
"loss": 6.14,
|
|
"mean_token_accuracy": 0.15635560899972917,
|
|
"num_tokens": 4108414.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.473322486877441,
|
|
"epoch": 1.9462827675118177,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004844824359623014,
|
|
"loss": 6.2727,
|
|
"mean_token_accuracy": 0.14365370869636535,
|
|
"num_tokens": 4117731.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.42958698272705,
|
|
"epoch": 1.9505801461108723,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048436087273952966,
|
|
"loss": 6.2456,
|
|
"mean_token_accuracy": 0.14321349710226058,
|
|
"num_tokens": 4127194.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.395676183700561,
|
|
"epoch": 1.9548775247099268,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048423885235629265,
|
|
"loss": 6.2082,
|
|
"mean_token_accuracy": 0.1566771924495697,
|
|
"num_tokens": 4135594.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.3766546726226805,
|
|
"epoch": 1.9591749033089814,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004841163750790342,
|
|
"loss": 6.2387,
|
|
"mean_token_accuracy": 0.1482451669871807,
|
|
"num_tokens": 4145027.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.411153888702392,
|
|
"epoch": 1.9634722819080361,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048399344117519555,
|
|
"loss": 6.0918,
|
|
"mean_token_accuracy": 0.15616220831871033,
|
|
"num_tokens": 4153754.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.349991703033448,
|
|
"epoch": 1.9677696605070907,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048387005091321544,
|
|
"loss": 6.1779,
|
|
"mean_token_accuracy": 0.15535371601581574,
|
|
"num_tokens": 4162765.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.42883620262146,
|
|
"epoch": 1.9720670391061452,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048374620456252877,
|
|
"loss": 6.1885,
|
|
"mean_token_accuracy": 0.1536056511104107,
|
|
"num_tokens": 4171589.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.385624837875366,
|
|
"epoch": 1.9763644177052,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048362190239356644,
|
|
"loss": 6.1974,
|
|
"mean_token_accuracy": 0.15068082362413407,
|
|
"num_tokens": 4181817.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.353290271759033,
|
|
"epoch": 1.9806617963042545,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048349714467775474,
|
|
"loss": 6.1594,
|
|
"mean_token_accuracy": 0.1520315356552601,
|
|
"num_tokens": 4191350.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.391085815429688,
|
|
"epoch": 1.984959174903309,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048337193168751464,
|
|
"loss": 6.2041,
|
|
"mean_token_accuracy": 0.1448718011379242,
|
|
"num_tokens": 4199888.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.419165563583374,
|
|
"epoch": 1.9892565535023636,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004832462636962613,
|
|
"loss": 6.1675,
|
|
"mean_token_accuracy": 0.15215626806020738,
|
|
"num_tokens": 4209509.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.369716596603394,
|
|
"epoch": 1.9935539321014182,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004831201409784034,
|
|
"loss": 6.1396,
|
|
"mean_token_accuracy": 0.154638309776783,
|
|
"num_tokens": 4218496.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.423878908157349,
|
|
"epoch": 1.9978513107004727,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004829935638093424,
|
|
"loss": 6.1503,
|
|
"mean_token_accuracy": 0.154294952750206,
|
|
"num_tokens": 4227504.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.347577359941271,
|
|
"epoch": 2.0017189514396216,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004828665324654724,
|
|
"loss": 6.0631,
|
|
"mean_token_accuracy": 0.15955217679341635,
|
|
"num_tokens": 4235338.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.4371147632598875,
|
|
"epoch": 2.006016330038676,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004827390472241791,
|
|
"loss": 5.8518,
|
|
"mean_token_accuracy": 0.1650273993611336,
|
|
"num_tokens": 4244905.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.327361726760865,
|
|
"epoch": 2.010313708637731,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004826111083638392,
|
|
"loss": 5.9297,
|
|
"mean_token_accuracy": 0.16078748032450677,
|
|
"num_tokens": 4254533.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.365427732467651,
|
|
"epoch": 2.0146110872367857,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048248271616382,
|
|
"loss": 5.8909,
|
|
"mean_token_accuracy": 0.16574428230524063,
|
|
"num_tokens": 4264023.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.342710208892822,
|
|
"epoch": 2.0189084658358403,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048235387090447894,
|
|
"loss": 5.9334,
|
|
"mean_token_accuracy": 0.16069018244743347,
|
|
"num_tokens": 4273298.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.364804077148437,
|
|
"epoch": 2.023205844434895,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048222457286716235,
|
|
"loss": 5.8764,
|
|
"mean_token_accuracy": 0.1686857357621193,
|
|
"num_tokens": 4283244.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.2878947257995605,
|
|
"epoch": 2.0275032230339494,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048209482233420564,
|
|
"loss": 5.8282,
|
|
"mean_token_accuracy": 0.1770843207836151,
|
|
"num_tokens": 4291677.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.3038060665130615,
|
|
"epoch": 2.031800601633004,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000481964619588932,
|
|
"loss": 5.8865,
|
|
"mean_token_accuracy": 0.16635389626026154,
|
|
"num_tokens": 4300822.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.329894876480102,
|
|
"epoch": 2.0360979802320585,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004818339649156523,
|
|
"loss": 5.9031,
|
|
"mean_token_accuracy": 0.16931673139333725,
|
|
"num_tokens": 4310149.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.202824544906616,
|
|
"epoch": 2.040395358831113,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048170285859966395,
|
|
"loss": 5.8119,
|
|
"mean_token_accuracy": 0.1753219798207283,
|
|
"num_tokens": 4319109.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.33490400314331,
|
|
"epoch": 2.0446927374301676,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048157130092725087,
|
|
"loss": 5.7905,
|
|
"mean_token_accuracy": 0.17381954342126846,
|
|
"num_tokens": 4327921.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.351769590377808,
|
|
"epoch": 2.048990116029222,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004814392921856824,
|
|
"loss": 5.9396,
|
|
"mean_token_accuracy": 0.16599524021148682,
|
|
"num_tokens": 4338026.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.274995040893555,
|
|
"epoch": 2.0532874946282766,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004813068326632128,
|
|
"loss": 5.7941,
|
|
"mean_token_accuracy": 0.1739266350865364,
|
|
"num_tokens": 4347794.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.335586738586426,
|
|
"epoch": 2.057584873227331,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004811739226490809,
|
|
"loss": 5.9723,
|
|
"mean_token_accuracy": 0.1657353535294533,
|
|
"num_tokens": 4357249.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.2787346839904785,
|
|
"epoch": 2.0618822518263857,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048104056243350896,
|
|
"loss": 5.9151,
|
|
"mean_token_accuracy": 0.1622727930545807,
|
|
"num_tokens": 4366053.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.314475345611572,
|
|
"epoch": 2.0661796304254403,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004809067523077023,
|
|
"loss": 5.9368,
|
|
"mean_token_accuracy": 0.16686248034238815,
|
|
"num_tokens": 4375543.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.281144666671753,
|
|
"epoch": 2.0704770090244953,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048077249256384884,
|
|
"loss": 5.8062,
|
|
"mean_token_accuracy": 0.17436677366495132,
|
|
"num_tokens": 4384332.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.224628639221192,
|
|
"epoch": 2.07477438762355,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004806377834951182,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.16276209503412248,
|
|
"num_tokens": 4393670.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.381396484375,
|
|
"epoch": 2.0790717662226044,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048050262539566104,
|
|
"loss": 5.9074,
|
|
"mean_token_accuracy": 0.1704580768942833,
|
|
"num_tokens": 4402763.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.253005933761597,
|
|
"epoch": 2.083369144821659,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004803670185606087,
|
|
"loss": 5.8316,
|
|
"mean_token_accuracy": 0.17470465898513793,
|
|
"num_tokens": 4411863.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.290292263031006,
|
|
"epoch": 2.0876665234207135,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004802309632860724,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.16799798905849456,
|
|
"num_tokens": 4421110.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.328047180175782,
|
|
"epoch": 2.091963902019768,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048009445986914236,
|
|
"loss": 5.9037,
|
|
"mean_token_accuracy": 0.16361760944128037,
|
|
"num_tokens": 4430249.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.251105260848999,
|
|
"epoch": 2.0962612806188226,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047995750860788756,
|
|
"loss": 5.8763,
|
|
"mean_token_accuracy": 0.16064124852418898,
|
|
"num_tokens": 4439686.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.269289493560791,
|
|
"epoch": 2.100558659217877,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004798201098013547,
|
|
"loss": 5.8784,
|
|
"mean_token_accuracy": 0.17077162116765976,
|
|
"num_tokens": 4448645.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.216794633865357,
|
|
"epoch": 2.1048560378169316,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047968226374956797,
|
|
"loss": 5.8564,
|
|
"mean_token_accuracy": 0.16725987941026688,
|
|
"num_tokens": 4456870.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.243628740310669,
|
|
"epoch": 2.109153416415986,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047954397075352794,
|
|
"loss": 5.8711,
|
|
"mean_token_accuracy": 0.16993128061294555,
|
|
"num_tokens": 4466287.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.267875003814697,
|
|
"epoch": 2.1134507950150407,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047940523111521136,
|
|
"loss": 5.7674,
|
|
"mean_token_accuracy": 0.17551021575927733,
|
|
"num_tokens": 4474461.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.209959268569946,
|
|
"epoch": 2.1177481736140953,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004792660451375701,
|
|
"loss": 5.8378,
|
|
"mean_token_accuracy": 0.1700508713722229,
|
|
"num_tokens": 4483002.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.252925443649292,
|
|
"epoch": 2.12204555221315,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047912641312453064,
|
|
"loss": 5.8445,
|
|
"mean_token_accuracy": 0.16888150274753572,
|
|
"num_tokens": 4492061.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.2727385520935055,
|
|
"epoch": 2.1263429308122044,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047898633538099363,
|
|
"loss": 5.9077,
|
|
"mean_token_accuracy": 0.16311377584934234,
|
|
"num_tokens": 4501829.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.240498495101929,
|
|
"epoch": 2.130640309411259,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004788458122128327,
|
|
"loss": 5.926,
|
|
"mean_token_accuracy": 0.16604274660348892,
|
|
"num_tokens": 4511539.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.272474479675293,
|
|
"epoch": 2.134937688010314,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047870484392689434,
|
|
"loss": 5.7914,
|
|
"mean_token_accuracy": 0.1672845244407654,
|
|
"num_tokens": 4520425.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.2719934463500975,
|
|
"epoch": 2.1392350666093685,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000478563430830997,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.16633028984069825,
|
|
"num_tokens": 4529474.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.207609510421753,
|
|
"epoch": 2.143532445208423,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047842157323393035,
|
|
"loss": 5.8296,
|
|
"mean_token_accuracy": 0.1706731289625168,
|
|
"num_tokens": 4538082.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.286458158493042,
|
|
"epoch": 2.1478298238074776,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004782792714454547,
|
|
"loss": 6.0015,
|
|
"mean_token_accuracy": 0.16442835330963135,
|
|
"num_tokens": 4547340.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.1478298238074776,
|
|
"eval_entropy": 6.109945370270325,
|
|
"eval_loss": 6.212513446807861,
|
|
"eval_mean_token_accuracy": 0.15539610120761502,
|
|
"eval_num_tokens": 4547340.0,
|
|
"eval_runtime": 2.0504,
|
|
"eval_samples_per_second": 1730.921,
|
|
"eval_steps_per_second": 216.548,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.2429650783538815,
|
|
"epoch": 2.152127202406532,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004781365257763002,
|
|
"loss": 5.8456,
|
|
"mean_token_accuracy": 0.16931284964084625,
|
|
"num_tokens": 4556415.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.164100790023804,
|
|
"epoch": 2.1564245810055866,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047799333653816633,
|
|
"loss": 5.7458,
|
|
"mean_token_accuracy": 0.1765029177069664,
|
|
"num_tokens": 4565156.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.2381916522979735,
|
|
"epoch": 2.160721959604641,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047784970404372124,
|
|
"loss": 5.8464,
|
|
"mean_token_accuracy": 0.16380368024110795,
|
|
"num_tokens": 4574678.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.172519302368164,
|
|
"epoch": 2.1650193382036957,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047770562860660083,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.16628692597150802,
|
|
"num_tokens": 4583253.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.255974197387696,
|
|
"epoch": 2.1693167168027503,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004775611105414083,
|
|
"loss": 5.9376,
|
|
"mean_token_accuracy": 0.1599043294787407,
|
|
"num_tokens": 4594042.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.199368286132812,
|
|
"epoch": 2.173614095401805,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004774161501637133,
|
|
"loss": 5.8834,
|
|
"mean_token_accuracy": 0.1693019151687622,
|
|
"num_tokens": 4603128.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.238662958145142,
|
|
"epoch": 2.1779114740008594,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004772707477900514,
|
|
"loss": 5.8539,
|
|
"mean_token_accuracy": 0.16817047595977783,
|
|
"num_tokens": 4611537.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.345929908752441,
|
|
"epoch": 2.182208852599914,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004771249037379232,
|
|
"loss": 5.968,
|
|
"mean_token_accuracy": 0.16030408143997193,
|
|
"num_tokens": 4622481.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.233392953872681,
|
|
"epoch": 2.1865062311989685,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004769786183257939,
|
|
"loss": 5.8709,
|
|
"mean_token_accuracy": 0.17347402423620223,
|
|
"num_tokens": 4631259.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.184610652923584,
|
|
"epoch": 2.190803609798023,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004768318918730924,
|
|
"loss": 5.7984,
|
|
"mean_token_accuracy": 0.17670518308877944,
|
|
"num_tokens": 4640266.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.218139505386352,
|
|
"epoch": 2.195100988397078,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047668472470021044,
|
|
"loss": 5.8713,
|
|
"mean_token_accuracy": 0.16367379873991011,
|
|
"num_tokens": 4649520.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.239636707305908,
|
|
"epoch": 2.1993983669961326,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004765371171285025,
|
|
"loss": 5.8163,
|
|
"mean_token_accuracy": 0.17504389137029647,
|
|
"num_tokens": 4658501.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.096165037155151,
|
|
"epoch": 2.203695745595187,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047638906948028445,
|
|
"loss": 5.8729,
|
|
"mean_token_accuracy": 0.16270434856414795,
|
|
"num_tokens": 4667567.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.269977855682373,
|
|
"epoch": 2.2079931241942417,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047624058207883317,
|
|
"loss": 5.8604,
|
|
"mean_token_accuracy": 0.17035209834575654,
|
|
"num_tokens": 4676618.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.308097791671753,
|
|
"epoch": 2.212290502793296,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047609165524838576,
|
|
"loss": 5.9352,
|
|
"mean_token_accuracy": 0.1621068611741066,
|
|
"num_tokens": 4685967.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.127339649200439,
|
|
"epoch": 2.2165878813923507,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004759422893141389,
|
|
"loss": 5.8236,
|
|
"mean_token_accuracy": 0.1692316710948944,
|
|
"num_tokens": 4694568.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.216268539428711,
|
|
"epoch": 2.2208852599914053,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004757924846022482,
|
|
"loss": 5.89,
|
|
"mean_token_accuracy": 0.1616070792078972,
|
|
"num_tokens": 4703648.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.20239634513855,
|
|
"epoch": 2.22518263859046,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047564224143982714,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.17824842780828476,
|
|
"num_tokens": 4712444.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.250929498672486,
|
|
"epoch": 2.2294800171895144,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047549156015494676,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.16583409011363984,
|
|
"num_tokens": 4722034.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.108888292312622,
|
|
"epoch": 2.233777395788569,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047534044107663484,
|
|
"loss": 5.9105,
|
|
"mean_token_accuracy": 0.1638724222779274,
|
|
"num_tokens": 4731344.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.311524677276611,
|
|
"epoch": 2.2380747743876235,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047518888453487496,
|
|
"loss": 5.817,
|
|
"mean_token_accuracy": 0.1758144274353981,
|
|
"num_tokens": 4739302.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.171684503555298,
|
|
"epoch": 2.242372152986678,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004750368908606061,
|
|
"loss": 5.9379,
|
|
"mean_token_accuracy": 0.16401576846837998,
|
|
"num_tokens": 4748848.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.278733396530152,
|
|
"epoch": 2.2466695315857326,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047488446038572164,
|
|
"loss": 5.988,
|
|
"mean_token_accuracy": 0.158454030752182,
|
|
"num_tokens": 4758194.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.271483707427978,
|
|
"epoch": 2.250966910184787,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004747315934430688,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.16533373892307282,
|
|
"num_tokens": 4768081.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.122837829589844,
|
|
"epoch": 2.2552642887838417,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000474578290366448,
|
|
"loss": 5.8258,
|
|
"mean_token_accuracy": 0.16424619704484938,
|
|
"num_tokens": 4776471.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.242071962356567,
|
|
"epoch": 2.259561667382896,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004744245514906117,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.17140334099531174,
|
|
"num_tokens": 4784403.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.109820699691772,
|
|
"epoch": 2.263859045981951,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047427037715126426,
|
|
"loss": 5.8071,
|
|
"mean_token_accuracy": 0.16793861836194993,
|
|
"num_tokens": 4792779.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.13071641921997,
|
|
"epoch": 2.2681564245810057,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004741157676850608,
|
|
"loss": 5.7959,
|
|
"mean_token_accuracy": 0.1719123378396034,
|
|
"num_tokens": 4801426.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.219759845733643,
|
|
"epoch": 2.2724538031800603,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047396072342960663,
|
|
"loss": 5.8436,
|
|
"mean_token_accuracy": 0.16473171710968018,
|
|
"num_tokens": 4810329.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.172075414657593,
|
|
"epoch": 2.276751181779115,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047380524472345645,
|
|
"loss": 5.8926,
|
|
"mean_token_accuracy": 0.16456550508737564,
|
|
"num_tokens": 4819544.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.204276752471924,
|
|
"epoch": 2.2810485603781694,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004736493319061134,
|
|
"loss": 5.9046,
|
|
"mean_token_accuracy": 0.16881225854158402,
|
|
"num_tokens": 4828113.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.19776291847229,
|
|
"epoch": 2.285345938977224,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004734929853180291,
|
|
"loss": 5.8878,
|
|
"mean_token_accuracy": 0.16559435576200485,
|
|
"num_tokens": 4836989.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.220638990402222,
|
|
"epoch": 2.2896433175762785,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00047333620530060175,
|
|
"loss": 5.9317,
|
|
"mean_token_accuracy": 0.1637803792953491,
|
|
"num_tokens": 4847103.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.251469230651855,
|
|
"epoch": 2.293940696175333,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004731789921961764,
|
|
"loss": 5.9358,
|
|
"mean_token_accuracy": 0.164635768532753,
|
|
"num_tokens": 4856238.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.212742853164673,
|
|
"epoch": 2.2982380747743876,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004730213463480434,
|
|
"loss": 5.8428,
|
|
"mean_token_accuracy": 0.17302587628364563,
|
|
"num_tokens": 4864608.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.159517288208008,
|
|
"epoch": 2.302535453373442,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047286326810043857,
|
|
"loss": 5.7962,
|
|
"mean_token_accuracy": 0.16983986347913743,
|
|
"num_tokens": 4873889.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.198332214355469,
|
|
"epoch": 2.3068328319724967,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047270475779854137,
|
|
"loss": 5.829,
|
|
"mean_token_accuracy": 0.1737384930253029,
|
|
"num_tokens": 4882902.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.295279121398925,
|
|
"epoch": 2.311130210571551,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047254581578847507,
|
|
"loss": 5.8707,
|
|
"mean_token_accuracy": 0.16589304506778718,
|
|
"num_tokens": 4892390.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.145449829101563,
|
|
"epoch": 2.3154275891706058,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004723864424173055,
|
|
"loss": 5.9872,
|
|
"mean_token_accuracy": 0.1639627918601036,
|
|
"num_tokens": 4901625.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.2244713306427,
|
|
"epoch": 2.3197249677696608,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004722266380330403,
|
|
"loss": 5.7716,
|
|
"mean_token_accuracy": 0.17777814120054244,
|
|
"num_tokens": 4910804.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.205074501037598,
|
|
"epoch": 2.3240223463687153,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047206640298462857,
|
|
"loss": 5.8554,
|
|
"mean_token_accuracy": 0.17233493477106093,
|
|
"num_tokens": 4920441.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.170949125289917,
|
|
"epoch": 2.32831972496777,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047190573762195945,
|
|
"loss": 5.8997,
|
|
"mean_token_accuracy": 0.1642414078116417,
|
|
"num_tokens": 4930204.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.167499303817749,
|
|
"epoch": 2.3326171035668244,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00047174464229586186,
|
|
"loss": 6.007,
|
|
"mean_token_accuracy": 0.15407418310642243,
|
|
"num_tokens": 4941191.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.3167987823486325,
|
|
"epoch": 2.336914482165879,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004715831173581036,
|
|
"loss": 5.9632,
|
|
"mean_token_accuracy": 0.16583859026432038,
|
|
"num_tokens": 4951825.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.154605293273926,
|
|
"epoch": 2.3412118607649335,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047142116316139073,
|
|
"loss": 5.9062,
|
|
"mean_token_accuracy": 0.16728972047567367,
|
|
"num_tokens": 4960632.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.263886308670044,
|
|
"epoch": 2.345509239363988,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004712587800593663,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.16565471589565278,
|
|
"num_tokens": 4969455.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.191297912597657,
|
|
"epoch": 2.3498066179630426,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004710959684066102,
|
|
"loss": 5.8388,
|
|
"mean_token_accuracy": 0.17293576747179032,
|
|
"num_tokens": 4978997.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.195657110214233,
|
|
"epoch": 2.354103996562097,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047093272855863803,
|
|
"loss": 5.9072,
|
|
"mean_token_accuracy": 0.16806580722332,
|
|
"num_tokens": 4988305.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.159595584869384,
|
|
"epoch": 2.3584013751611517,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004707690608719003,
|
|
"loss": 5.8381,
|
|
"mean_token_accuracy": 0.17171266674995422,
|
|
"num_tokens": 4997022.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.204063940048218,
|
|
"epoch": 2.362698753760206,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004706049657037818,
|
|
"loss": 5.8984,
|
|
"mean_token_accuracy": 0.16405170410871506,
|
|
"num_tokens": 5005664.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.20027117729187,
|
|
"epoch": 2.3669961323592608,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004704404434126009,
|
|
"loss": 5.8461,
|
|
"mean_token_accuracy": 0.16249971985816955,
|
|
"num_tokens": 5014769.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.224915170669556,
|
|
"epoch": 2.3712935109583153,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047027549435760843,
|
|
"loss": 5.9216,
|
|
"mean_token_accuracy": 0.1661619648337364,
|
|
"num_tokens": 5024060.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.301250553131103,
|
|
"epoch": 2.37559088955737,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004701101188989872,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.15823127031326295,
|
|
"num_tokens": 5033046.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.196396017074585,
|
|
"epoch": 2.3798882681564244,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046994431739785114,
|
|
"loss": 5.8026,
|
|
"mean_token_accuracy": 0.1763622686266899,
|
|
"num_tokens": 5040894.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.190552377700806,
|
|
"epoch": 2.384185646755479,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046977809021624454,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.16489340364933014,
|
|
"num_tokens": 5050961.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.236513423919678,
|
|
"epoch": 2.3884830253545335,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004696114377171409,
|
|
"loss": 5.9022,
|
|
"mean_token_accuracy": 0.1675196796655655,
|
|
"num_tokens": 5060226.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.134419250488281,
|
|
"epoch": 2.3927804039535885,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004694443602644429,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.1658073604106903,
|
|
"num_tokens": 5069225.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.230590677261352,
|
|
"epoch": 2.397077782552643,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004692768582229808,
|
|
"loss": 5.8537,
|
|
"mean_token_accuracy": 0.16909998953342437,
|
|
"num_tokens": 5078386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.158745527267456,
|
|
"epoch": 2.4013751611516976,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046910893195851213,
|
|
"loss": 5.7812,
|
|
"mean_token_accuracy": 0.16957359611988068,
|
|
"num_tokens": 5087161.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.144252824783325,
|
|
"epoch": 2.405672539750752,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046894058183772074,
|
|
"loss": 5.9448,
|
|
"mean_token_accuracy": 0.16215559244155883,
|
|
"num_tokens": 5096613.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.177387857437134,
|
|
"epoch": 2.4099699183498067,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000468771808228216,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.16563187092542647,
|
|
"num_tokens": 5106534.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.154799270629883,
|
|
"epoch": 2.414267296948861,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046860261149853197,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.16382392197847367,
|
|
"num_tokens": 5115975.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.148438119888306,
|
|
"epoch": 2.4185646755479158,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004684329920181268,
|
|
"loss": 5.8216,
|
|
"mean_token_accuracy": 0.17049064636230468,
|
|
"num_tokens": 5124635.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.1651019096374515,
|
|
"epoch": 2.4228620541469703,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046826295015738154,
|
|
"loss": 5.7785,
|
|
"mean_token_accuracy": 0.17763484865427018,
|
|
"num_tokens": 5133226.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.099349737167358,
|
|
"epoch": 2.427159432746025,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004680924862875996,
|
|
"loss": 5.8788,
|
|
"mean_token_accuracy": 0.170813749730587,
|
|
"num_tokens": 5142257.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.231349277496338,
|
|
"epoch": 2.4314568113450794,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046792160078100605,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.16604618430137635,
|
|
"num_tokens": 5150752.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.136086511611938,
|
|
"epoch": 2.435754189944134,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046775029401074653,
|
|
"loss": 5.7844,
|
|
"mean_token_accuracy": 0.178993821144104,
|
|
"num_tokens": 5160237.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.202201557159424,
|
|
"epoch": 2.4400515685431885,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046757856635088645,
|
|
"loss": 5.8537,
|
|
"mean_token_accuracy": 0.1712223544716835,
|
|
"num_tokens": 5169752.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.172966527938843,
|
|
"epoch": 2.444348947142243,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004674064181764105,
|
|
"loss": 5.9056,
|
|
"mean_token_accuracy": 0.1733956053853035,
|
|
"num_tokens": 5178892.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.177707719802856,
|
|
"epoch": 2.448646325741298,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046723384986322147,
|
|
"loss": 5.8777,
|
|
"mean_token_accuracy": 0.16896003782749175,
|
|
"num_tokens": 5188468.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.167446327209473,
|
|
"epoch": 2.4529437043403526,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004670608617881395,
|
|
"loss": 5.8142,
|
|
"mean_token_accuracy": 0.17425985783338546,
|
|
"num_tokens": 5197565.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.108637952804566,
|
|
"epoch": 2.457241082939407,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004668874543289014,
|
|
"loss": 5.7986,
|
|
"mean_token_accuracy": 0.18018365651369095,
|
|
"num_tokens": 5205791.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.141421508789063,
|
|
"epoch": 2.4615384615384617,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046671362786415986,
|
|
"loss": 5.8011,
|
|
"mean_token_accuracy": 0.18125482499599457,
|
|
"num_tokens": 5214773.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.088820695877075,
|
|
"epoch": 2.465835840137516,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046653938277348237,
|
|
"loss": 5.8487,
|
|
"mean_token_accuracy": 0.17472687363624573,
|
|
"num_tokens": 5223734.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.261274719238282,
|
|
"epoch": 2.4701332187365708,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004663647194373505,
|
|
"loss": 5.9156,
|
|
"mean_token_accuracy": 0.16283378154039382,
|
|
"num_tokens": 5231742.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.124586343765259,
|
|
"epoch": 2.4744305973356253,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046618963823715913,
|
|
"loss": 5.8653,
|
|
"mean_token_accuracy": 0.1686026558279991,
|
|
"num_tokens": 5241673.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.198731851577759,
|
|
"epoch": 2.47872797593468,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046601413955521575,
|
|
"loss": 5.8336,
|
|
"mean_token_accuracy": 0.1733954593539238,
|
|
"num_tokens": 5250082.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.167904710769653,
|
|
"epoch": 2.4830253545337344,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004658382237747393,
|
|
"loss": 5.894,
|
|
"mean_token_accuracy": 0.16233221739530562,
|
|
"num_tokens": 5259680.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.210194540023804,
|
|
"epoch": 2.487322733132789,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046566189127985946,
|
|
"loss": 5.8836,
|
|
"mean_token_accuracy": 0.17433235198259353,
|
|
"num_tokens": 5269561.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.169331502914429,
|
|
"epoch": 2.4916201117318435,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000465485142455616,
|
|
"loss": 5.8388,
|
|
"mean_token_accuracy": 0.1740201309323311,
|
|
"num_tokens": 5278659.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.07768177986145,
|
|
"epoch": 2.495917490330898,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046530797768795765,
|
|
"loss": 5.801,
|
|
"mean_token_accuracy": 0.17944221645593644,
|
|
"num_tokens": 5287619.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.17024712562561,
|
|
"epoch": 2.5002148689299526,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046513039736374153,
|
|
"loss": 5.9382,
|
|
"mean_token_accuracy": 0.16346622705459596,
|
|
"num_tokens": 5297334.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.215419769287109,
|
|
"epoch": 2.504512247529007,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004649524018707319,
|
|
"loss": 5.8424,
|
|
"mean_token_accuracy": 0.16681308597326278,
|
|
"num_tokens": 5306208.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.142744636535644,
|
|
"epoch": 2.5088096261280617,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046477399159759996,
|
|
"loss": 5.7948,
|
|
"mean_token_accuracy": 0.17883567363023758,
|
|
"num_tokens": 5314754.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.037441778182983,
|
|
"epoch": 2.5131070047271162,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046459516693392246,
|
|
"loss": 5.8052,
|
|
"mean_token_accuracy": 0.17655965834856033,
|
|
"num_tokens": 5324000.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.213449382781983,
|
|
"epoch": 2.517404383326171,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004644159282701808,
|
|
"loss": 5.8495,
|
|
"mean_token_accuracy": 0.16930186077952386,
|
|
"num_tokens": 5332478.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.226251840591431,
|
|
"epoch": 2.5217017619252258,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00046423627599776076,
|
|
"loss": 5.9213,
|
|
"mean_token_accuracy": 0.15651765167713166,
|
|
"num_tokens": 5341635.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.143741273880005,
|
|
"epoch": 2.5259991405242803,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000464056210508951,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.16428655087947847,
|
|
"num_tokens": 5350144.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.176555585861206,
|
|
"epoch": 2.530296519123335,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004638757321969426,
|
|
"loss": 5.8319,
|
|
"mean_token_accuracy": 0.16960337311029433,
|
|
"num_tokens": 5358788.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.169870615005493,
|
|
"epoch": 2.5345938977223894,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046369484145582815,
|
|
"loss": 5.9296,
|
|
"mean_token_accuracy": 0.16125051379203797,
|
|
"num_tokens": 5368057.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.0688502311706545,
|
|
"epoch": 2.538891276321444,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046351353868060054,
|
|
"loss": 5.7804,
|
|
"mean_token_accuracy": 0.17502057552337646,
|
|
"num_tokens": 5376739.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.1720726013183596,
|
|
"epoch": 2.5431886549204985,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046333182426715273,
|
|
"loss": 5.8841,
|
|
"mean_token_accuracy": 0.16769777685403825,
|
|
"num_tokens": 5385967.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.232366323471069,
|
|
"epoch": 2.547486033519553,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046314969861227626,
|
|
"loss": 5.9187,
|
|
"mean_token_accuracy": 0.16190613955259323,
|
|
"num_tokens": 5395192.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.1295037269592285,
|
|
"epoch": 2.5517834121186076,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004629671621136608,
|
|
"loss": 5.876,
|
|
"mean_token_accuracy": 0.17085160166025162,
|
|
"num_tokens": 5404694.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.156940174102783,
|
|
"epoch": 2.556080790717662,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004627842151698931,
|
|
"loss": 5.8791,
|
|
"mean_token_accuracy": 0.16839216351509095,
|
|
"num_tokens": 5413102.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.141358661651611,
|
|
"epoch": 2.5603781693167167,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046260085818045625,
|
|
"loss": 5.9057,
|
|
"mean_token_accuracy": 0.16761772185564042,
|
|
"num_tokens": 5423339.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.211736726760864,
|
|
"epoch": 2.5646755479157712,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004624170915457284,
|
|
"loss": 5.8683,
|
|
"mean_token_accuracy": 0.17020333409309388,
|
|
"num_tokens": 5432377.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.145984411239624,
|
|
"epoch": 2.5689729265148262,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046223291566698264,
|
|
"loss": 5.787,
|
|
"mean_token_accuracy": 0.17505069077014923,
|
|
"num_tokens": 5441038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.096357202529907,
|
|
"epoch": 2.5732703051138808,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004620483309463855,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.17221614569425583,
|
|
"num_tokens": 5449557.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.217533254623413,
|
|
"epoch": 2.5775676837129353,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004618633377869961,
|
|
"loss": 5.9446,
|
|
"mean_token_accuracy": 0.15857684314250947,
|
|
"num_tokens": 5458931.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.5775676837129353,
|
|
"eval_entropy": 6.002345256977253,
|
|
"eval_loss": 6.1212897300720215,
|
|
"eval_mean_token_accuracy": 0.16277532707396392,
|
|
"eval_num_tokens": 5458931.0,
|
|
"eval_runtime": 2.0558,
|
|
"eval_samples_per_second": 1726.341,
|
|
"eval_steps_per_second": 215.975,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.1347846508026125,
|
|
"epoch": 2.58186506231199,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004616779365927656,
|
|
"loss": 5.7734,
|
|
"mean_token_accuracy": 0.1822523444890976,
|
|
"num_tokens": 5468539.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.024096488952637,
|
|
"epoch": 2.5861624409110444,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004614921277685361,
|
|
"loss": 5.7085,
|
|
"mean_token_accuracy": 0.17936478108167647,
|
|
"num_tokens": 5475710.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.1361793041229244,
|
|
"epoch": 2.590459819510099,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046130591172003976,
|
|
"loss": 5.8527,
|
|
"mean_token_accuracy": 0.17134219259023667,
|
|
"num_tokens": 5484597.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.120989418029785,
|
|
"epoch": 2.5947571981091535,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004611192888538981,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.1608140453696251,
|
|
"num_tokens": 5493213.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.2168285846710205,
|
|
"epoch": 2.599054576708208,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046093225957762084,
|
|
"loss": 5.9308,
|
|
"mean_token_accuracy": 0.16593084782361983,
|
|
"num_tokens": 5502556.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.197446489334107,
|
|
"epoch": 2.6033519553072626,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004607448242996051,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.17084067910909653,
|
|
"num_tokens": 5511779.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.1454957008361815,
|
|
"epoch": 2.607649333906317,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004605569834291347,
|
|
"loss": 5.8161,
|
|
"mean_token_accuracy": 0.17526462078094482,
|
|
"num_tokens": 5520836.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.083783388137817,
|
|
"epoch": 2.6119467125053717,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046036873737637904,
|
|
"loss": 5.8457,
|
|
"mean_token_accuracy": 0.17209848016500473,
|
|
"num_tokens": 5529285.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.101385116577148,
|
|
"epoch": 2.6162440911044262,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004601800865523921,
|
|
"loss": 5.8519,
|
|
"mean_token_accuracy": 0.16869369894266129,
|
|
"num_tokens": 5538160.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.142069292068482,
|
|
"epoch": 2.620541469703481,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045999103136911204,
|
|
"loss": 5.8559,
|
|
"mean_token_accuracy": 0.16590654104948044,
|
|
"num_tokens": 5547355.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.143155193328857,
|
|
"epoch": 2.6248388483025353,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045980157223935965,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.17009032517671585,
|
|
"num_tokens": 5557299.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.0996771335601805,
|
|
"epoch": 2.62913622690159,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045961170957683806,
|
|
"loss": 5.7957,
|
|
"mean_token_accuracy": 0.1701590210199356,
|
|
"num_tokens": 5565469.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.1573389053344725,
|
|
"epoch": 2.6334336055006444,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045942144379613147,
|
|
"loss": 5.9159,
|
|
"mean_token_accuracy": 0.1658922776579857,
|
|
"num_tokens": 5574740.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.166552925109864,
|
|
"epoch": 2.637730984099699,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045923077531270426,
|
|
"loss": 5.8854,
|
|
"mean_token_accuracy": 0.16845837086439133,
|
|
"num_tokens": 5583438.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.17145938873291,
|
|
"epoch": 2.6420283626987535,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004590397045429001,
|
|
"loss": 5.857,
|
|
"mean_token_accuracy": 0.17479295134544373,
|
|
"num_tokens": 5592389.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.079704666137696,
|
|
"epoch": 2.646325741297808,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045884823190394134,
|
|
"loss": 5.7566,
|
|
"mean_token_accuracy": 0.1772332966327667,
|
|
"num_tokens": 5601598.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.118323564529419,
|
|
"epoch": 2.650623119896863,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004586563578139275,
|
|
"loss": 5.8446,
|
|
"mean_token_accuracy": 0.16692771315574645,
|
|
"num_tokens": 5610498.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.109970378875732,
|
|
"epoch": 2.6549204984959176,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045846408269183505,
|
|
"loss": 5.7654,
|
|
"mean_token_accuracy": 0.1782268226146698,
|
|
"num_tokens": 5620082.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.152325868606567,
|
|
"epoch": 2.659217877094972,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045827140695751603,
|
|
"loss": 5.8366,
|
|
"mean_token_accuracy": 0.1716820165514946,
|
|
"num_tokens": 5630291.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.097200155258179,
|
|
"epoch": 2.6635152556940267,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004580783310316971,
|
|
"loss": 5.8345,
|
|
"mean_token_accuracy": 0.17504153102636338,
|
|
"num_tokens": 5638784.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.027546215057373,
|
|
"epoch": 2.6678126342930812,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045788485533597895,
|
|
"loss": 5.6879,
|
|
"mean_token_accuracy": 0.18603624403476715,
|
|
"num_tokens": 5647968.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.150567054748535,
|
|
"epoch": 2.672110012892136,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045769098029283526,
|
|
"loss": 5.9347,
|
|
"mean_token_accuracy": 0.16201953887939452,
|
|
"num_tokens": 5657543.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.15388331413269,
|
|
"epoch": 2.6764073914911903,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004574967063256115,
|
|
"loss": 5.8353,
|
|
"mean_token_accuracy": 0.17699069976806642,
|
|
"num_tokens": 5666535.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.084828329086304,
|
|
"epoch": 2.680704770090245,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045730203385852447,
|
|
"loss": 5.9196,
|
|
"mean_token_accuracy": 0.16706804931163788,
|
|
"num_tokens": 5676273.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.041564130783081,
|
|
"epoch": 2.6850021486892994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000457106963316661,
|
|
"loss": 5.8215,
|
|
"mean_token_accuracy": 0.17681703120470046,
|
|
"num_tokens": 5684888.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.1651557922363285,
|
|
"epoch": 2.689299527288354,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045691149512597717,
|
|
"loss": 5.8817,
|
|
"mean_token_accuracy": 0.16579522490501403,
|
|
"num_tokens": 5693626.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.171338653564453,
|
|
"epoch": 2.6935969058874085,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00045671562971329736,
|
|
"loss": 5.7997,
|
|
"mean_token_accuracy": 0.17142064571380616,
|
|
"num_tokens": 5702542.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.036250162124634,
|
|
"epoch": 2.6978942844864635,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045651936750631337,
|
|
"loss": 5.8332,
|
|
"mean_token_accuracy": 0.17233078181743622,
|
|
"num_tokens": 5711440.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.216920471191406,
|
|
"epoch": 2.702191663085518,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045632270893358333,
|
|
"loss": 5.8818,
|
|
"mean_token_accuracy": 0.16918585151433946,
|
|
"num_tokens": 5721495.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.1993273258209225,
|
|
"epoch": 2.7064890416845726,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004561256544245312,
|
|
"loss": 5.9189,
|
|
"mean_token_accuracy": 0.1607155531644821,
|
|
"num_tokens": 5730664.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.047014474868774,
|
|
"epoch": 2.710786420283627,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000455928204409445,
|
|
"loss": 5.8044,
|
|
"mean_token_accuracy": 0.17973511815071105,
|
|
"num_tokens": 5740229.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.113893842697143,
|
|
"epoch": 2.7150837988826817,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045573035931947684,
|
|
"loss": 5.802,
|
|
"mean_token_accuracy": 0.17425440698862077,
|
|
"num_tokens": 5748549.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.111674499511719,
|
|
"epoch": 2.7193811774817362,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004555321195866411,
|
|
"loss": 5.7391,
|
|
"mean_token_accuracy": 0.17396723330020905,
|
|
"num_tokens": 5757603.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.137631750106811,
|
|
"epoch": 2.723678556080791,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004553334856438143,
|
|
"loss": 5.9224,
|
|
"mean_token_accuracy": 0.16666966825723648,
|
|
"num_tokens": 5767520.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.1654016971588135,
|
|
"epoch": 2.7279759346798453,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00045513445792473356,
|
|
"loss": 5.911,
|
|
"mean_token_accuracy": 0.16542696654796601,
|
|
"num_tokens": 5776778.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.174024152755737,
|
|
"epoch": 2.7322733132789,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004549350368639958,
|
|
"loss": 5.9451,
|
|
"mean_token_accuracy": 0.16035651415586472,
|
|
"num_tokens": 5785652.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.2061694145202635,
|
|
"epoch": 2.7365706918779544,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045473522289705693,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.1694064900279045,
|
|
"num_tokens": 5795766.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.0452875137329105,
|
|
"epoch": 2.740868070477009,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045453501646023085,
|
|
"loss": 5.9372,
|
|
"mean_token_accuracy": 0.16121204793453217,
|
|
"num_tokens": 5804504.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.124906301498413,
|
|
"epoch": 2.7451654490760635,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045433441799068837,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.17111808210611343,
|
|
"num_tokens": 5814161.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.147184801101685,
|
|
"epoch": 2.749462827675118,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004541334279264562,
|
|
"loss": 5.7407,
|
|
"mean_token_accuracy": 0.18340681195259095,
|
|
"num_tokens": 5822235.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.072987508773804,
|
|
"epoch": 2.7537602062741726,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045393204670641656,
|
|
"loss": 5.7547,
|
|
"mean_token_accuracy": 0.1745822474360466,
|
|
"num_tokens": 5831572.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.005992603302002,
|
|
"epoch": 2.758057584873227,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004537302747703055,
|
|
"loss": 5.7829,
|
|
"mean_token_accuracy": 0.17934428602457048,
|
|
"num_tokens": 5839694.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.185523939132691,
|
|
"epoch": 2.7623549634722817,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045352811255871216,
|
|
"loss": 5.9086,
|
|
"mean_token_accuracy": 0.16498518288135527,
|
|
"num_tokens": 5849131.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.20290732383728,
|
|
"epoch": 2.7666523420713363,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00045332556051307804,
|
|
"loss": 5.8267,
|
|
"mean_token_accuracy": 0.16665343046188355,
|
|
"num_tokens": 5858861.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.123448705673217,
|
|
"epoch": 2.770949720670391,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045312261907569585,
|
|
"loss": 5.8263,
|
|
"mean_token_accuracy": 0.17210643440485002,
|
|
"num_tokens": 5867585.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.09133882522583,
|
|
"epoch": 2.775247099269446,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045291928868970867,
|
|
"loss": 5.8557,
|
|
"mean_token_accuracy": 0.16730334907770156,
|
|
"num_tokens": 5876256.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.103731679916382,
|
|
"epoch": 2.7795444778685003,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004527155697991087,
|
|
"loss": 5.9124,
|
|
"mean_token_accuracy": 0.16498502045869828,
|
|
"num_tokens": 5885302.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.116323661804199,
|
|
"epoch": 2.783841856467555,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004525114628487365,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.1643183633685112,
|
|
"num_tokens": 5895066.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.18259449005127,
|
|
"epoch": 2.7881392350666094,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045230696828428026,
|
|
"loss": 5.9034,
|
|
"mean_token_accuracy": 0.16625797897577285,
|
|
"num_tokens": 5903258.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.126916885375977,
|
|
"epoch": 2.792436613665664,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004521020865522742,
|
|
"loss": 5.7972,
|
|
"mean_token_accuracy": 0.171038818359375,
|
|
"num_tokens": 5911714.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.091484212875367,
|
|
"epoch": 2.7967339922647185,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045189681810009827,
|
|
"loss": 5.8604,
|
|
"mean_token_accuracy": 0.1673845484852791,
|
|
"num_tokens": 5920432.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.245140838623047,
|
|
"epoch": 2.801031370863773,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045169116337597653,
|
|
"loss": 5.8792,
|
|
"mean_token_accuracy": 0.16876531690359114,
|
|
"num_tokens": 5929202.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.171068525314331,
|
|
"epoch": 2.8053287494628276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000451485122828977,
|
|
"loss": 5.9096,
|
|
"mean_token_accuracy": 0.1670824408531189,
|
|
"num_tokens": 5938034.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.007605361938476,
|
|
"epoch": 2.809626128061882,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045127869690900956,
|
|
"loss": 5.7687,
|
|
"mean_token_accuracy": 0.17781976610422134,
|
|
"num_tokens": 5946944.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.040731048583984,
|
|
"epoch": 2.8139235066609367,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045107188606682613,
|
|
"loss": 5.8725,
|
|
"mean_token_accuracy": 0.17112763077020646,
|
|
"num_tokens": 5956475.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.213654851913452,
|
|
"epoch": 2.8182208852599913,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004508646907540188,
|
|
"loss": 5.8253,
|
|
"mean_token_accuracy": 0.16726296544075012,
|
|
"num_tokens": 5965814.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.115692520141602,
|
|
"epoch": 2.8225182638590463,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004506571114230195,
|
|
"loss": 5.8797,
|
|
"mean_token_accuracy": 0.16483918130397796,
|
|
"num_tokens": 5973850.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.044183731079102,
|
|
"epoch": 2.826815642458101,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045044914852709824,
|
|
"loss": 5.8337,
|
|
"mean_token_accuracy": 0.16715002954006195,
|
|
"num_tokens": 5982987.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.176396608352661,
|
|
"epoch": 2.8311130210571553,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004502408025203631,
|
|
"loss": 5.8088,
|
|
"mean_token_accuracy": 0.17391630709171296,
|
|
"num_tokens": 5992227.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.108564949035644,
|
|
"epoch": 2.83541039965621,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004500320738577584,
|
|
"loss": 5.8006,
|
|
"mean_token_accuracy": 0.17633765786886216,
|
|
"num_tokens": 6000243.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.053935432434082,
|
|
"epoch": 2.8397077782552644,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044982296299506407,
|
|
"loss": 5.7997,
|
|
"mean_token_accuracy": 0.17893847674131394,
|
|
"num_tokens": 6009771.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.1164305210113525,
|
|
"epoch": 2.844005156854319,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004496134703888948,
|
|
"loss": 5.875,
|
|
"mean_token_accuracy": 0.1690862849354744,
|
|
"num_tokens": 6018683.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.122239971160889,
|
|
"epoch": 2.8483025354533735,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044940359649669846,
|
|
"loss": 5.7361,
|
|
"mean_token_accuracy": 0.18076154738664627,
|
|
"num_tokens": 6027422.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.093537425994873,
|
|
"epoch": 2.852599914052428,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044919334177675595,
|
|
"loss": 5.8232,
|
|
"mean_token_accuracy": 0.16410151720046998,
|
|
"num_tokens": 6035670.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.088628053665161,
|
|
"epoch": 2.8568972926514826,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044898270668817955,
|
|
"loss": 5.7633,
|
|
"mean_token_accuracy": 0.177126844227314,
|
|
"num_tokens": 6044092.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.090199422836304,
|
|
"epoch": 2.861194671250537,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000448771691690912,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.16831086575984955,
|
|
"num_tokens": 6053970.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.140215587615967,
|
|
"epoch": 2.8654920498495917,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004485602972457257,
|
|
"loss": 5.7898,
|
|
"mean_token_accuracy": 0.1754717782139778,
|
|
"num_tokens": 6062965.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.150375843048096,
|
|
"epoch": 2.8697894284486463,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044834852381422165,
|
|
"loss": 5.8695,
|
|
"mean_token_accuracy": 0.17155641317367554,
|
|
"num_tokens": 6072420.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.0596644401550295,
|
|
"epoch": 2.874086807047701,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044813637185882836,
|
|
"loss": 5.7734,
|
|
"mean_token_accuracy": 0.1735791265964508,
|
|
"num_tokens": 6080915.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.116873121261596,
|
|
"epoch": 2.8783841856467554,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044792384184280106,
|
|
"loss": 5.8954,
|
|
"mean_token_accuracy": 0.16623952835798264,
|
|
"num_tokens": 6090453.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.090284633636474,
|
|
"epoch": 2.88268156424581,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044771093423022013,
|
|
"loss": 5.9298,
|
|
"mean_token_accuracy": 0.15913961827754974,
|
|
"num_tokens": 6099390.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.114023637771607,
|
|
"epoch": 2.8869789428448644,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004474976494859909,
|
|
"loss": 5.8486,
|
|
"mean_token_accuracy": 0.17675736397504807,
|
|
"num_tokens": 6108677.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.098732328414917,
|
|
"epoch": 2.891276321443919,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004472839880758419,
|
|
"loss": 5.7711,
|
|
"mean_token_accuracy": 0.17180029749870301,
|
|
"num_tokens": 6117151.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.195354080200195,
|
|
"epoch": 2.8955737000429735,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004470699504663242,
|
|
"loss": 5.8942,
|
|
"mean_token_accuracy": 0.16719547659158707,
|
|
"num_tokens": 6127167.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.082501411437988,
|
|
"epoch": 2.899871078642028,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004468555371248104,
|
|
"loss": 5.7776,
|
|
"mean_token_accuracy": 0.17165370136499405,
|
|
"num_tokens": 6136487.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.114503192901611,
|
|
"epoch": 2.904168457241083,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004466407485194937,
|
|
"loss": 5.8846,
|
|
"mean_token_accuracy": 0.16588175743818284,
|
|
"num_tokens": 6145334.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.101830673217774,
|
|
"epoch": 2.9084658358401376,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004464255851193864,
|
|
"loss": 5.8134,
|
|
"mean_token_accuracy": 0.16928868889808654,
|
|
"num_tokens": 6155062.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.060112714767456,
|
|
"epoch": 2.912763214439192,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0004462100473943194,
|
|
"loss": 5.779,
|
|
"mean_token_accuracy": 0.17902038842439652,
|
|
"num_tokens": 6164313.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.066328811645508,
|
|
"epoch": 2.9170605930382467,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000445994135814941,
|
|
"loss": 5.8031,
|
|
"mean_token_accuracy": 0.17068170607089997,
|
|
"num_tokens": 6173513.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.1107556343078615,
|
|
"epoch": 2.9213579716373013,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044577785085271566,
|
|
"loss": 5.8222,
|
|
"mean_token_accuracy": 0.16928994953632354,
|
|
"num_tokens": 6182000.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.177258014678955,
|
|
"epoch": 2.925655350236356,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004455611929799235,
|
|
"loss": 5.8653,
|
|
"mean_token_accuracy": 0.15872504711151122,
|
|
"num_tokens": 6191887.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.03788800239563,
|
|
"epoch": 2.9299527288354104,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004453441626696585,
|
|
"loss": 5.8994,
|
|
"mean_token_accuracy": 0.15877071171998977,
|
|
"num_tokens": 6202897.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.142374467849732,
|
|
"epoch": 2.934250107434465,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044512676039582823,
|
|
"loss": 5.7983,
|
|
"mean_token_accuracy": 0.17737043499946595,
|
|
"num_tokens": 6211811.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.133689117431641,
|
|
"epoch": 2.9385474860335195,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004449089866331524,
|
|
"loss": 5.807,
|
|
"mean_token_accuracy": 0.17649049013853074,
|
|
"num_tokens": 6219896.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.010902643203735,
|
|
"epoch": 2.942844864632574,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004446908418571617,
|
|
"loss": 5.7923,
|
|
"mean_token_accuracy": 0.17107276171445845,
|
|
"num_tokens": 6228212.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.108887815475464,
|
|
"epoch": 2.9471422432316285,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004444723265441973,
|
|
"loss": 5.9385,
|
|
"mean_token_accuracy": 0.16253667175769806,
|
|
"num_tokens": 6238133.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.079803562164306,
|
|
"epoch": 2.9514396218306835,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004442534411714092,
|
|
"loss": 5.8493,
|
|
"mean_token_accuracy": 0.1624876856803894,
|
|
"num_tokens": 6247331.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.170568227767944,
|
|
"epoch": 2.955737000429738,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044403418621675555,
|
|
"loss": 5.8456,
|
|
"mean_token_accuracy": 0.1654752418398857,
|
|
"num_tokens": 6255280.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.140178108215332,
|
|
"epoch": 2.9600343790287926,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004438145621590017,
|
|
"loss": 5.8073,
|
|
"mean_token_accuracy": 0.17526279389858246,
|
|
"num_tokens": 6264752.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.028224134445191,
|
|
"epoch": 2.964331757627847,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044359456947771857,
|
|
"loss": 5.7614,
|
|
"mean_token_accuracy": 0.1661372035741806,
|
|
"num_tokens": 6273258.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.952836751937866,
|
|
"epoch": 2.9686291362269017,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004433742086532824,
|
|
"loss": 5.6712,
|
|
"mean_token_accuracy": 0.18851299732923507,
|
|
"num_tokens": 6281584.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.098459720611572,
|
|
"epoch": 2.9729265148259563,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044315348016687317,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.17163048237562178,
|
|
"num_tokens": 6290016.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.039625978469848,
|
|
"epoch": 2.977223893425011,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004429323845004736,
|
|
"loss": 5.7066,
|
|
"mean_token_accuracy": 0.17723663598299028,
|
|
"num_tokens": 6298569.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.064740991592407,
|
|
"epoch": 2.9815212720240654,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00044271092213686824,
|
|
"loss": 5.7313,
|
|
"mean_token_accuracy": 0.17582760155200958,
|
|
"num_tokens": 6307684.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.119738626480102,
|
|
"epoch": 2.98581865062312,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00044248909355964247,
|
|
"loss": 5.8843,
|
|
"mean_token_accuracy": 0.16779844611883163,
|
|
"num_tokens": 6317767.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.1805627822875975,
|
|
"epoch": 2.9901160292221745,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044226689925318117,
|
|
"loss": 5.8958,
|
|
"mean_token_accuracy": 0.16598885059356688,
|
|
"num_tokens": 6327457.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.034802579879761,
|
|
"epoch": 2.994413407821229,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044204433970266785,
|
|
"loss": 5.6993,
|
|
"mean_token_accuracy": 0.18884851187467575,
|
|
"num_tokens": 6335747.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.0412510395050045,
|
|
"epoch": 2.9987107864202835,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004418214153940837,
|
|
"loss": 5.7915,
|
|
"mean_token_accuracy": 0.1731286823749542,
|
|
"num_tokens": 6344750.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.094671514299181,
|
|
"epoch": 3.002578427159433,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00044159812681420624,
|
|
"loss": 5.7345,
|
|
"mean_token_accuracy": 0.1751474059290356,
|
|
"num_tokens": 6354779.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.1476243019104,
|
|
"epoch": 3.0068758057584875,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004413744744506086,
|
|
"loss": 5.5077,
|
|
"mean_token_accuracy": 0.1835872769355774,
|
|
"num_tokens": 6363809.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.0068758057584875,
|
|
"eval_entropy": 5.855848801028621,
|
|
"eval_loss": 6.045755863189697,
|
|
"eval_mean_token_accuracy": 0.16792396449290956,
|
|
"eval_num_tokens": 6363809.0,
|
|
"eval_runtime": 2.0563,
|
|
"eval_samples_per_second": 1725.906,
|
|
"eval_steps_per_second": 215.921,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11630,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1434986114734080.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|