7537dbe9df
Model: fpadovani/eus-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.2887838418564677,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 613354283642880.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|