7537dbe9df
Model: fpadovani/eus-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.42973785990545765,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 204362498211840.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|