35fe8e09a3
Model: fpadovani/eus-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.4366136656639448,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.6312949657440186,
|
|
"epoch": 0.004297378599054577,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.384,
|
|
"mean_token_accuracy": 0.09047168418765068,
|
|
"num_tokens": 10107.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.674387979507446,
|
|
"epoch": 0.008594757198109154,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.3814,
|
|
"mean_token_accuracy": 0.09915048182010651,
|
|
"num_tokens": 18391.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.658490705490112,
|
|
"epoch": 0.01289213579716373,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.4194,
|
|
"mean_token_accuracy": 0.09372682273387908,
|
|
"num_tokens": 27061.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.6485553741455075,
|
|
"epoch": 0.017189514396218308,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.4387,
|
|
"mean_token_accuracy": 0.09950413554906845,
|
|
"num_tokens": 36339.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.655299663543701,
|
|
"epoch": 0.021486892995272882,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.4336,
|
|
"mean_token_accuracy": 0.09199422970414162,
|
|
"num_tokens": 45770.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.707321071624756,
|
|
"epoch": 0.02578427159432746,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.4406,
|
|
"mean_token_accuracy": 0.09267855286598206,
|
|
"num_tokens": 54575.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.718957376480103,
|
|
"epoch": 0.030081650193382038,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.5222,
|
|
"mean_token_accuracy": 0.08976790606975556,
|
|
"num_tokens": 66403.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.742082262039185,
|
|
"epoch": 0.034379028792436615,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.4377,
|
|
"mean_token_accuracy": 0.09164252653717994,
|
|
"num_tokens": 76510.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.745701646804809,
|
|
"epoch": 0.03867640739149119,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.358,
|
|
"mean_token_accuracy": 0.0955798089504242,
|
|
"num_tokens": 84836.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.780595874786377,
|
|
"epoch": 0.042973785990545764,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.3289,
|
|
"mean_token_accuracy": 0.10552914068102837,
|
|
"num_tokens": 93197.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.764179325103759,
|
|
"epoch": 0.047271164589600345,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.3234,
|
|
"mean_token_accuracy": 0.09917277097702026,
|
|
"num_tokens": 101546.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.719727945327759,
|
|
"epoch": 0.05156854318865492,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.4172,
|
|
"mean_token_accuracy": 0.0928034670650959,
|
|
"num_tokens": 111703.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.748228645324707,
|
|
"epoch": 0.055865921787709494,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.3403,
|
|
"mean_token_accuracy": 0.10037123262882233,
|
|
"num_tokens": 119894.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.714352416992187,
|
|
"epoch": 0.060163300386764075,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.2915,
|
|
"mean_token_accuracy": 0.1022428810596466,
|
|
"num_tokens": 128885.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.679376173019409,
|
|
"epoch": 0.06446067898581866,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.4226,
|
|
"mean_token_accuracy": 0.0972097434103489,
|
|
"num_tokens": 138106.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.72790002822876,
|
|
"epoch": 0.06875805758487323,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.3294,
|
|
"mean_token_accuracy": 0.1022751808166504,
|
|
"num_tokens": 146691.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.730126142501831,
|
|
"epoch": 0.0730554361839278,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.382,
|
|
"mean_token_accuracy": 0.09973402544856072,
|
|
"num_tokens": 155792.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.727601718902588,
|
|
"epoch": 0.07735281478298238,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.4474,
|
|
"mean_token_accuracy": 0.08758748695254326,
|
|
"num_tokens": 166944.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.782265329360962,
|
|
"epoch": 0.08165019338203695,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.2886,
|
|
"mean_token_accuracy": 0.1041356198489666,
|
|
"num_tokens": 175303.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.751953029632569,
|
|
"epoch": 0.08594757198109153,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.3403,
|
|
"mean_token_accuracy": 0.09793160557746887,
|
|
"num_tokens": 184708.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.702822208404541,
|
|
"epoch": 0.09024495058014612,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.3117,
|
|
"mean_token_accuracy": 0.09851032048463822,
|
|
"num_tokens": 193835.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.686660861968994,
|
|
"epoch": 0.09454232917920069,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.3479,
|
|
"mean_token_accuracy": 0.0979080393910408,
|
|
"num_tokens": 203344.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.698584461212159,
|
|
"epoch": 0.09883970777825526,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.4586,
|
|
"mean_token_accuracy": 0.09130895733833314,
|
|
"num_tokens": 213048.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.781258678436279,
|
|
"epoch": 0.10313708637730984,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.3094,
|
|
"mean_token_accuracy": 0.10353164449334144,
|
|
"num_tokens": 221784.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.650211572647095,
|
|
"epoch": 0.10743446497636441,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.3189,
|
|
"mean_token_accuracy": 0.09726176261901856,
|
|
"num_tokens": 230971.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.655170726776123,
|
|
"epoch": 0.11173184357541899,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.2818,
|
|
"mean_token_accuracy": 0.1042576052248478,
|
|
"num_tokens": 240524.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.7341550350189205,
|
|
"epoch": 0.11602922217447358,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.2512,
|
|
"mean_token_accuracy": 0.1007460281252861,
|
|
"num_tokens": 249220.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.745693302154541,
|
|
"epoch": 0.12032660077352815,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.3688,
|
|
"mean_token_accuracy": 0.10030856803059578,
|
|
"num_tokens": 258934.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.694993305206299,
|
|
"epoch": 0.12462397937258272,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.2936,
|
|
"mean_token_accuracy": 0.10321335718035698,
|
|
"num_tokens": 267680.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.719129991531372,
|
|
"epoch": 0.1289213579716373,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.3236,
|
|
"mean_token_accuracy": 0.10207543894648552,
|
|
"num_tokens": 276227.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.648375129699707,
|
|
"epoch": 0.1332187365706919,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.2203,
|
|
"mean_token_accuracy": 0.1059327855706215,
|
|
"num_tokens": 286342.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.674158382415771,
|
|
"epoch": 0.13751611516974646,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.2988,
|
|
"mean_token_accuracy": 0.09665355160832405,
|
|
"num_tokens": 294994.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.717900514602661,
|
|
"epoch": 0.14181349376880104,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.2704,
|
|
"mean_token_accuracy": 0.10349940955638885,
|
|
"num_tokens": 303882.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.6729988098144535,
|
|
"epoch": 0.1461108723678556,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.3104,
|
|
"mean_token_accuracy": 0.10128599181771278,
|
|
"num_tokens": 312515.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.739007139205933,
|
|
"epoch": 0.15040825096691018,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.27,
|
|
"mean_token_accuracy": 0.10081852003931999,
|
|
"num_tokens": 320801.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.720875406265259,
|
|
"epoch": 0.15470562956596476,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2872,
|
|
"mean_token_accuracy": 0.10100285485386848,
|
|
"num_tokens": 329382.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.66646089553833,
|
|
"epoch": 0.15900300816501933,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.2814,
|
|
"mean_token_accuracy": 0.1028428927063942,
|
|
"num_tokens": 337894.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.772510719299317,
|
|
"epoch": 0.1633003867640739,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2803,
|
|
"mean_token_accuracy": 0.10378619506955147,
|
|
"num_tokens": 346380.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.690706968307495,
|
|
"epoch": 0.16759776536312848,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3588,
|
|
"mean_token_accuracy": 0.09733301475644111,
|
|
"num_tokens": 356305.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.79454927444458,
|
|
"epoch": 0.17189514396218306,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.306,
|
|
"mean_token_accuracy": 0.09683404862880707,
|
|
"num_tokens": 364899.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.694888687133789,
|
|
"epoch": 0.17619252256123766,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2938,
|
|
"mean_token_accuracy": 0.09810400977730752,
|
|
"num_tokens": 373663.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.748025798797608,
|
|
"epoch": 0.18048990116029223,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2566,
|
|
"mean_token_accuracy": 0.10043591782450675,
|
|
"num_tokens": 382730.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.706165361404419,
|
|
"epoch": 0.1847872797593468,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3157,
|
|
"mean_token_accuracy": 0.09612104147672654,
|
|
"num_tokens": 392676.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.760982656478882,
|
|
"epoch": 0.18908465835840138,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2955,
|
|
"mean_token_accuracy": 0.10281639397144318,
|
|
"num_tokens": 401050.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.626513719558716,
|
|
"epoch": 0.19338203695745596,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2692,
|
|
"mean_token_accuracy": 0.10119878426194191,
|
|
"num_tokens": 410009.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.726489019393921,
|
|
"epoch": 0.19767941555651053,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2683,
|
|
"mean_token_accuracy": 0.10186234638094901,
|
|
"num_tokens": 419302.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.643717670440674,
|
|
"epoch": 0.2019767941555651,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1665,
|
|
"mean_token_accuracy": 0.10647615045309067,
|
|
"num_tokens": 427296.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.666737127304077,
|
|
"epoch": 0.20627417275461968,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.3139,
|
|
"mean_token_accuracy": 0.10131902173161507,
|
|
"num_tokens": 436368.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.772911167144775,
|
|
"epoch": 0.21057155135367425,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2112,
|
|
"mean_token_accuracy": 0.1055280588567257,
|
|
"num_tokens": 445535.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.602903366088867,
|
|
"epoch": 0.21486892995272883,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2153,
|
|
"mean_token_accuracy": 0.10406075567007064,
|
|
"num_tokens": 454769.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.693030595779419,
|
|
"epoch": 0.2191663085517834,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.2315,
|
|
"mean_token_accuracy": 0.10270996242761612,
|
|
"num_tokens": 463975.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.637308835983276,
|
|
"epoch": 0.22346368715083798,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.2542,
|
|
"mean_token_accuracy": 0.10225536078214645,
|
|
"num_tokens": 472899.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.740519666671753,
|
|
"epoch": 0.22776106574989258,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.229,
|
|
"mean_token_accuracy": 0.1005932256579399,
|
|
"num_tokens": 481556.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.654651689529419,
|
|
"epoch": 0.23205844434894715,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2258,
|
|
"mean_token_accuracy": 0.10702893435955048,
|
|
"num_tokens": 490253.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.660864973068238,
|
|
"epoch": 0.23635582294800173,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2451,
|
|
"mean_token_accuracy": 0.10333684608340263,
|
|
"num_tokens": 498444.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.637535953521729,
|
|
"epoch": 0.2406532015470563,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.191,
|
|
"mean_token_accuracy": 0.10794568434357643,
|
|
"num_tokens": 508330.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.6566917419433596,
|
|
"epoch": 0.24495058014611087,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3004,
|
|
"mean_token_accuracy": 0.10417937636375427,
|
|
"num_tokens": 517900.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.670303010940552,
|
|
"epoch": 0.24924795874516545,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2276,
|
|
"mean_token_accuracy": 0.10308908969163895,
|
|
"num_tokens": 527808.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.719700765609741,
|
|
"epoch": 0.25354533734422,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2415,
|
|
"mean_token_accuracy": 0.10010977610945701,
|
|
"num_tokens": 536931.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.668509387969971,
|
|
"epoch": 0.2578427159432746,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.279,
|
|
"mean_token_accuracy": 0.10248880609869956,
|
|
"num_tokens": 545758.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.700217819213867,
|
|
"epoch": 0.26214009454232917,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2819,
|
|
"mean_token_accuracy": 0.10198702886700631,
|
|
"num_tokens": 555165.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.6267822265625,
|
|
"epoch": 0.2664374731413838,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2035,
|
|
"mean_token_accuracy": 0.10117841735482216,
|
|
"num_tokens": 564719.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.646708202362061,
|
|
"epoch": 0.2707348517404383,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.1638,
|
|
"mean_token_accuracy": 0.10670615658164025,
|
|
"num_tokens": 573572.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.759027910232544,
|
|
"epoch": 0.2750322303394929,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.3476,
|
|
"mean_token_accuracy": 0.10210367739200592,
|
|
"num_tokens": 581497.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.590592908859253,
|
|
"epoch": 0.27932960893854747,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2138,
|
|
"mean_token_accuracy": 0.10664469674229622,
|
|
"num_tokens": 591107.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.70356388092041,
|
|
"epoch": 0.28362698753760207,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2482,
|
|
"mean_token_accuracy": 0.1050640620291233,
|
|
"num_tokens": 600241.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.639587259292602,
|
|
"epoch": 0.2879243661366566,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.161,
|
|
"mean_token_accuracy": 0.1065776713192463,
|
|
"num_tokens": 608697.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.602131795883179,
|
|
"epoch": 0.2922217447357112,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1698,
|
|
"mean_token_accuracy": 0.1098954938352108,
|
|
"num_tokens": 617275.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.669042348861694,
|
|
"epoch": 0.29651912333476577,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.2602,
|
|
"mean_token_accuracy": 0.1007254920899868,
|
|
"num_tokens": 626644.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.623440217971802,
|
|
"epoch": 0.30081650193382037,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1639,
|
|
"mean_token_accuracy": 0.1080157920718193,
|
|
"num_tokens": 635110.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.711002826690674,
|
|
"epoch": 0.30511388053287497,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.3139,
|
|
"mean_token_accuracy": 0.10216462090611458,
|
|
"num_tokens": 644746.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.708708238601685,
|
|
"epoch": 0.3094112591319295,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2216,
|
|
"mean_token_accuracy": 0.1021303728222847,
|
|
"num_tokens": 654281.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.534019136428833,
|
|
"epoch": 0.3137086377309841,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.2333,
|
|
"mean_token_accuracy": 0.10576817691326142,
|
|
"num_tokens": 663174.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.660452365875244,
|
|
"epoch": 0.31800601633003867,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.1525,
|
|
"mean_token_accuracy": 0.10541519671678543,
|
|
"num_tokens": 672178.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.651990938186645,
|
|
"epoch": 0.32230339492909327,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1748,
|
|
"mean_token_accuracy": 0.10421534106135369,
|
|
"num_tokens": 681323.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.537337684631348,
|
|
"epoch": 0.3266007735281478,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1001,
|
|
"mean_token_accuracy": 0.11140918657183647,
|
|
"num_tokens": 690461.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.596573305130005,
|
|
"epoch": 0.3308981521272024,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.1461,
|
|
"mean_token_accuracy": 0.10594902262091636,
|
|
"num_tokens": 699199.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.566946506500244,
|
|
"epoch": 0.33519553072625696,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.109,
|
|
"mean_token_accuracy": 0.11522968709468842,
|
|
"num_tokens": 707949.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.66830849647522,
|
|
"epoch": 0.33949290932531156,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1843,
|
|
"mean_token_accuracy": 0.10416831225156784,
|
|
"num_tokens": 715752.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.619978666305542,
|
|
"epoch": 0.3437902879243661,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1119,
|
|
"mean_token_accuracy": 0.11198346018791198,
|
|
"num_tokens": 724416.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.594716548919678,
|
|
"epoch": 0.3480876665234207,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.1774,
|
|
"mean_token_accuracy": 0.10296614542603492,
|
|
"num_tokens": 733116.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.614369249343872,
|
|
"epoch": 0.3523850451224753,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.1639,
|
|
"mean_token_accuracy": 0.10737873241305351,
|
|
"num_tokens": 742093.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.532227945327759,
|
|
"epoch": 0.35668242372152986,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.1385,
|
|
"mean_token_accuracy": 0.11264142915606498,
|
|
"num_tokens": 750402.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.510246276855469,
|
|
"epoch": 0.36097980232058446,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.1129,
|
|
"mean_token_accuracy": 0.11108387559652329,
|
|
"num_tokens": 760961.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.720337963104248,
|
|
"epoch": 0.365277180919639,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.2042,
|
|
"mean_token_accuracy": 0.10612902790307999,
|
|
"num_tokens": 770554.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.437310361862183,
|
|
"epoch": 0.3695745595186936,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.1596,
|
|
"mean_token_accuracy": 0.11299800872802734,
|
|
"num_tokens": 779172.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.663910818099976,
|
|
"epoch": 0.37387193811774816,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.2239,
|
|
"mean_token_accuracy": 0.10290571823716163,
|
|
"num_tokens": 788040.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.589281415939331,
|
|
"epoch": 0.37816931671680276,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.1461,
|
|
"mean_token_accuracy": 0.10722599253058433,
|
|
"num_tokens": 796786.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.543337059020996,
|
|
"epoch": 0.3824666953158573,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.1192,
|
|
"mean_token_accuracy": 0.10885161831974983,
|
|
"num_tokens": 805520.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.486078453063965,
|
|
"epoch": 0.3867640739149119,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.074,
|
|
"mean_token_accuracy": 0.10658745989203453,
|
|
"num_tokens": 814939.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.534557342529297,
|
|
"epoch": 0.39106145251396646,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.0766,
|
|
"mean_token_accuracy": 0.11227057129144669,
|
|
"num_tokens": 823862.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.5476549625396725,
|
|
"epoch": 0.39535883111302106,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.1124,
|
|
"mean_token_accuracy": 0.10576009079813957,
|
|
"num_tokens": 832820.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.601094675064087,
|
|
"epoch": 0.39965620971207566,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0697,
|
|
"mean_token_accuracy": 0.11121490225195885,
|
|
"num_tokens": 841538.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.544060945510864,
|
|
"epoch": 0.4039535883111302,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.2069,
|
|
"mean_token_accuracy": 0.10181558132171631,
|
|
"num_tokens": 851123.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.549469089508056,
|
|
"epoch": 0.4082509669101848,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.1633,
|
|
"mean_token_accuracy": 0.11091246008872986,
|
|
"num_tokens": 860357.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.547894096374511,
|
|
"epoch": 0.41254834550923936,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.0874,
|
|
"mean_token_accuracy": 0.10722309574484826,
|
|
"num_tokens": 869980.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.507503604888916,
|
|
"epoch": 0.41684572410829396,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.0572,
|
|
"mean_token_accuracy": 0.11242355704307556,
|
|
"num_tokens": 878250.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.5191121101379395,
|
|
"epoch": 0.4211431027073485,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.1411,
|
|
"mean_token_accuracy": 0.11158529818058013,
|
|
"num_tokens": 887624.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.454204320907593,
|
|
"epoch": 0.4254404813064031,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.1159,
|
|
"mean_token_accuracy": 0.11260272860527039,
|
|
"num_tokens": 897120.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.495032835006714,
|
|
"epoch": 0.42973785990545765,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.0795,
|
|
"mean_token_accuracy": 0.11134620234370232,
|
|
"num_tokens": 906215.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.42973785990545765,
|
|
"eval_entropy": 7.203803374960616,
|
|
"eval_loss": 7.096514701843262,
|
|
"eval_mean_token_accuracy": 0.11462040213649874,
|
|
"eval_num_tokens": 906215.0,
|
|
"eval_runtime": 2.0645,
|
|
"eval_samples_per_second": 1719.022,
|
|
"eval_steps_per_second": 215.059,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.447824621200562,
|
|
"epoch": 0.43403523850451226,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.0811,
|
|
"mean_token_accuracy": 0.1122453585267067,
|
|
"num_tokens": 915181.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.498021125793457,
|
|
"epoch": 0.4383326171035668,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.1044,
|
|
"mean_token_accuracy": 0.10958386138081551,
|
|
"num_tokens": 924377.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.607626008987427,
|
|
"epoch": 0.4426299957026214,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.1944,
|
|
"mean_token_accuracy": 0.10655399709939957,
|
|
"num_tokens": 933114.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.6139122486114506,
|
|
"epoch": 0.44692737430167595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.1453,
|
|
"mean_token_accuracy": 0.11119715198874473,
|
|
"num_tokens": 943306.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.436026573181152,
|
|
"epoch": 0.45122475290073055,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.0354,
|
|
"mean_token_accuracy": 0.11904665902256965,
|
|
"num_tokens": 951515.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.494698238372803,
|
|
"epoch": 0.45552213149978515,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.1519,
|
|
"mean_token_accuracy": 0.10504961535334587,
|
|
"num_tokens": 962686.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.572213172912598,
|
|
"epoch": 0.4598195100988397,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.1449,
|
|
"mean_token_accuracy": 0.11348244249820709,
|
|
"num_tokens": 972136.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.405817127227783,
|
|
"epoch": 0.4641168886978943,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.0518,
|
|
"mean_token_accuracy": 0.1100372053682804,
|
|
"num_tokens": 981301.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.484500360488892,
|
|
"epoch": 0.46841426729694885,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.0823,
|
|
"mean_token_accuracy": 0.1120329774916172,
|
|
"num_tokens": 990360.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.573296546936035,
|
|
"epoch": 0.47271164589600345,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.1293,
|
|
"mean_token_accuracy": 0.10760239511728287,
|
|
"num_tokens": 999415.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.419287919998169,
|
|
"epoch": 0.477009024495058,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.057,
|
|
"mean_token_accuracy": 0.10999582111835479,
|
|
"num_tokens": 1008762.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.44342451095581,
|
|
"epoch": 0.4813064030941126,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.0505,
|
|
"mean_token_accuracy": 0.11702658385038375,
|
|
"num_tokens": 1017704.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.457871007919311,
|
|
"epoch": 0.48560378169316715,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.018,
|
|
"mean_token_accuracy": 0.11318592131137847,
|
|
"num_tokens": 1026251.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.356105470657349,
|
|
"epoch": 0.48990116029222175,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.0083,
|
|
"mean_token_accuracy": 0.11355392187833786,
|
|
"num_tokens": 1036191.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.5119133472442625,
|
|
"epoch": 0.4941985388912763,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.0501,
|
|
"mean_token_accuracy": 0.11168754398822785,
|
|
"num_tokens": 1044936.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.406773805618286,
|
|
"epoch": 0.4984959174903309,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.0476,
|
|
"mean_token_accuracy": 0.1135815680027008,
|
|
"num_tokens": 1053683.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.3828895568847654,
|
|
"epoch": 0.5027932960893855,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.0283,
|
|
"mean_token_accuracy": 0.11782724559307098,
|
|
"num_tokens": 1062932.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.4789910316467285,
|
|
"epoch": 0.50709067468844,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.0524,
|
|
"mean_token_accuracy": 0.11150057762861251,
|
|
"num_tokens": 1072313.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.458136653900146,
|
|
"epoch": 0.5113880532874946,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.033,
|
|
"mean_token_accuracy": 0.10738502442836761,
|
|
"num_tokens": 1081675.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.437460470199585,
|
|
"epoch": 0.5156854318865493,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.0392,
|
|
"mean_token_accuracy": 0.11078862249851226,
|
|
"num_tokens": 1091541.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.43347053527832,
|
|
"epoch": 0.5199828104856038,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.0467,
|
|
"mean_token_accuracy": 0.11545747444033623,
|
|
"num_tokens": 1100724.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.34070782661438,
|
|
"epoch": 0.5242801890846583,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.0062,
|
|
"mean_token_accuracy": 0.11681902781128883,
|
|
"num_tokens": 1108869.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.513333511352539,
|
|
"epoch": 0.5285775676837129,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.0303,
|
|
"mean_token_accuracy": 0.11391275599598885,
|
|
"num_tokens": 1117314.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.237616014480591,
|
|
"epoch": 0.5328749462827675,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.969,
|
|
"mean_token_accuracy": 0.11866867989301681,
|
|
"num_tokens": 1126786.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.403380393981934,
|
|
"epoch": 0.5371723248818221,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.983,
|
|
"mean_token_accuracy": 0.11322688534855843,
|
|
"num_tokens": 1136013.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.355997228622437,
|
|
"epoch": 0.5414697034808766,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.0163,
|
|
"mean_token_accuracy": 0.1159099243581295,
|
|
"num_tokens": 1144970.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.416441440582275,
|
|
"epoch": 0.5457670820799312,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.9784,
|
|
"mean_token_accuracy": 0.12343248203396798,
|
|
"num_tokens": 1153810.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.320913982391358,
|
|
"epoch": 0.5500644606789858,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.96,
|
|
"mean_token_accuracy": 0.11895549520850182,
|
|
"num_tokens": 1162498.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.383200359344483,
|
|
"epoch": 0.5543618392780404,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.0441,
|
|
"mean_token_accuracy": 0.11171148270368576,
|
|
"num_tokens": 1172091.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.465569925308228,
|
|
"epoch": 0.5586592178770949,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.0379,
|
|
"mean_token_accuracy": 0.1126454509794712,
|
|
"num_tokens": 1181400.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.29718279838562,
|
|
"epoch": 0.5629565964761496,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.0066,
|
|
"mean_token_accuracy": 0.11692977026104927,
|
|
"num_tokens": 1189780.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.376112461090088,
|
|
"epoch": 0.5672539750752041,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.9708,
|
|
"mean_token_accuracy": 0.11179102137684822,
|
|
"num_tokens": 1198671.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.406812715530395,
|
|
"epoch": 0.5715513536742587,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.9887,
|
|
"mean_token_accuracy": 0.11439693570137024,
|
|
"num_tokens": 1207173.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.267558336257935,
|
|
"epoch": 0.5758487322733132,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.9252,
|
|
"mean_token_accuracy": 0.11824023947119713,
|
|
"num_tokens": 1216387.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.466721105575561,
|
|
"epoch": 0.5801461108723679,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.9093,
|
|
"mean_token_accuracy": 0.11586858034133911,
|
|
"num_tokens": 1224461.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.260802936553955,
|
|
"epoch": 0.5844434894714224,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.9855,
|
|
"mean_token_accuracy": 0.1176436722278595,
|
|
"num_tokens": 1233774.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.267514610290528,
|
|
"epoch": 0.588740868070477,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.9319,
|
|
"mean_token_accuracy": 0.12313097864389419,
|
|
"num_tokens": 1242812.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.451924133300781,
|
|
"epoch": 0.5930382466695315,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.0445,
|
|
"mean_token_accuracy": 0.1125735655426979,
|
|
"num_tokens": 1252872.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.1216278076171875,
|
|
"epoch": 0.5973356252685862,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.8314,
|
|
"mean_token_accuracy": 0.1210754469037056,
|
|
"num_tokens": 1260852.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.292500305175781,
|
|
"epoch": 0.6016330038676407,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.9419,
|
|
"mean_token_accuracy": 0.1167706459760666,
|
|
"num_tokens": 1268925.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.384844732284546,
|
|
"epoch": 0.6059303824666953,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.9849,
|
|
"mean_token_accuracy": 0.11300796419382095,
|
|
"num_tokens": 1278994.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.286926889419556,
|
|
"epoch": 0.6102277610657499,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.9847,
|
|
"mean_token_accuracy": 0.11259545534849166,
|
|
"num_tokens": 1287698.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.337662601470948,
|
|
"epoch": 0.6145251396648045,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.9117,
|
|
"mean_token_accuracy": 0.12028303518891334,
|
|
"num_tokens": 1297475.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.265739297866821,
|
|
"epoch": 0.618822518263859,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.9558,
|
|
"mean_token_accuracy": 0.11790136769413948,
|
|
"num_tokens": 1306836.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.3774675846099855,
|
|
"epoch": 0.6231198968629136,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.9932,
|
|
"mean_token_accuracy": 0.11299360319972038,
|
|
"num_tokens": 1315872.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.3129335880279545,
|
|
"epoch": 0.6274172754619682,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.9353,
|
|
"mean_token_accuracy": 0.12453719973564148,
|
|
"num_tokens": 1324624.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.300215101242065,
|
|
"epoch": 0.6317146540610228,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.9246,
|
|
"mean_token_accuracy": 0.12120431885123253,
|
|
"num_tokens": 1333058.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.065497016906738,
|
|
"epoch": 0.6360120326600773,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.8904,
|
|
"mean_token_accuracy": 0.11625659838318825,
|
|
"num_tokens": 1342376.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.412401533126831,
|
|
"epoch": 0.6403094112591319,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.9293,
|
|
"mean_token_accuracy": 0.11268759667873382,
|
|
"num_tokens": 1351386.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.194233036041259,
|
|
"epoch": 0.6446067898581865,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.8338,
|
|
"mean_token_accuracy": 0.12849506586790085,
|
|
"num_tokens": 1358958.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.3347986221313475,
|
|
"epoch": 0.6489041684572411,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.988,
|
|
"mean_token_accuracy": 0.11507417485117913,
|
|
"num_tokens": 1368599.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.380126667022705,
|
|
"epoch": 0.6532015470562956,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.0127,
|
|
"mean_token_accuracy": 0.111283528059721,
|
|
"num_tokens": 1378529.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.157611989974976,
|
|
"epoch": 0.6574989256553503,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.8052,
|
|
"mean_token_accuracy": 0.1265752285718918,
|
|
"num_tokens": 1386993.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.21686282157898,
|
|
"epoch": 0.6617963042544048,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.8936,
|
|
"mean_token_accuracy": 0.12180712148547172,
|
|
"num_tokens": 1395790.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.166302919387817,
|
|
"epoch": 0.6660936828534594,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.9063,
|
|
"mean_token_accuracy": 0.11845313757658005,
|
|
"num_tokens": 1405587.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.20961365699768,
|
|
"epoch": 0.6703910614525139,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.8702,
|
|
"mean_token_accuracy": 0.12274195328354835,
|
|
"num_tokens": 1414478.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.319825458526611,
|
|
"epoch": 0.6746884400515686,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.9317,
|
|
"mean_token_accuracy": 0.12083822339773179,
|
|
"num_tokens": 1423791.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.313541460037231,
|
|
"epoch": 0.6789858186506231,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.975,
|
|
"mean_token_accuracy": 0.11185284182429314,
|
|
"num_tokens": 1432955.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.242367315292358,
|
|
"epoch": 0.6832831972496777,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.9394,
|
|
"mean_token_accuracy": 0.11529579535126686,
|
|
"num_tokens": 1441907.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.173644304275513,
|
|
"epoch": 0.6875805758487322,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.8059,
|
|
"mean_token_accuracy": 0.12198502644896507,
|
|
"num_tokens": 1451062.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.2840491771698,
|
|
"epoch": 0.6918779544477869,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.8894,
|
|
"mean_token_accuracy": 0.11644295528531075,
|
|
"num_tokens": 1460132.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.085446500778199,
|
|
"epoch": 0.6961753330468414,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7896,
|
|
"mean_token_accuracy": 0.12437586709856988,
|
|
"num_tokens": 1469582.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.180881690979004,
|
|
"epoch": 0.700472711645896,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.8844,
|
|
"mean_token_accuracy": 0.11694586053490638,
|
|
"num_tokens": 1479053.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.176044559478759,
|
|
"epoch": 0.7047700902449506,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.8874,
|
|
"mean_token_accuracy": 0.11812442615628242,
|
|
"num_tokens": 1488189.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.071721315383911,
|
|
"epoch": 0.7090674688440052,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.7495,
|
|
"mean_token_accuracy": 0.12273769155144691,
|
|
"num_tokens": 1497324.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.243275499343872,
|
|
"epoch": 0.7133648474430597,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.8631,
|
|
"mean_token_accuracy": 0.12297548577189446,
|
|
"num_tokens": 1506543.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.1102629661560055,
|
|
"epoch": 0.7176622260421143,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.8571,
|
|
"mean_token_accuracy": 0.1257997862994671,
|
|
"num_tokens": 1516737.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.015081739425659,
|
|
"epoch": 0.7219596046411689,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.7311,
|
|
"mean_token_accuracy": 0.12102818563580513,
|
|
"num_tokens": 1525561.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.17170901298523,
|
|
"epoch": 0.7262569832402235,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.757,
|
|
"mean_token_accuracy": 0.12571127861738204,
|
|
"num_tokens": 1533323.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.173940944671631,
|
|
"epoch": 0.730554361839278,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.821,
|
|
"mean_token_accuracy": 0.12750849053263663,
|
|
"num_tokens": 1542632.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.148316097259522,
|
|
"epoch": 0.7348517404383326,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.7649,
|
|
"mean_token_accuracy": 0.12507490813732147,
|
|
"num_tokens": 1551236.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.981910467147827,
|
|
"epoch": 0.7391491190373872,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.7641,
|
|
"mean_token_accuracy": 0.12514904662966728,
|
|
"num_tokens": 1559674.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.186282157897949,
|
|
"epoch": 0.7434464976364418,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.8498,
|
|
"mean_token_accuracy": 0.1250532478094101,
|
|
"num_tokens": 1569481.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.118600702285766,
|
|
"epoch": 0.7477438762354963,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8888,
|
|
"mean_token_accuracy": 0.1209896370768547,
|
|
"num_tokens": 1578488.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.105226039886475,
|
|
"epoch": 0.752041254834551,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.7736,
|
|
"mean_token_accuracy": 0.12527675032615662,
|
|
"num_tokens": 1586675.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.185068035125733,
|
|
"epoch": 0.7563386334336055,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.8782,
|
|
"mean_token_accuracy": 0.1180253192782402,
|
|
"num_tokens": 1595411.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.179415893554688,
|
|
"epoch": 0.7606360120326601,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8619,
|
|
"mean_token_accuracy": 0.12292847484350204,
|
|
"num_tokens": 1604046.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.130577564239502,
|
|
"epoch": 0.7649333906317146,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8566,
|
|
"mean_token_accuracy": 0.11715829819440841,
|
|
"num_tokens": 1613759.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.111226511001587,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.8191,
|
|
"mean_token_accuracy": 0.1252148814499378,
|
|
"num_tokens": 1623323.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.097943353652954,
|
|
"epoch": 0.7735281478298238,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.7922,
|
|
"mean_token_accuracy": 0.11943844705820084,
|
|
"num_tokens": 1631727.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.073408317565918,
|
|
"epoch": 0.7778255264288784,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.7454,
|
|
"mean_token_accuracy": 0.12582483813166617,
|
|
"num_tokens": 1639544.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.1905022144317625,
|
|
"epoch": 0.7821229050279329,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8716,
|
|
"mean_token_accuracy": 0.11673429310321808,
|
|
"num_tokens": 1648931.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.032827425003052,
|
|
"epoch": 0.7864202836269876,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.7325,
|
|
"mean_token_accuracy": 0.12737771049141883,
|
|
"num_tokens": 1657688.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.160619735717773,
|
|
"epoch": 0.7907176622260421,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.8191,
|
|
"mean_token_accuracy": 0.11969996094703675,
|
|
"num_tokens": 1666879.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.016655492782593,
|
|
"epoch": 0.7950150408250967,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.7912,
|
|
"mean_token_accuracy": 0.12404834032058716,
|
|
"num_tokens": 1676773.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.205742454528808,
|
|
"epoch": 0.7993124194241513,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.8942,
|
|
"mean_token_accuracy": 0.11682869419455529,
|
|
"num_tokens": 1686144.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.093483018875122,
|
|
"epoch": 0.8036097980232059,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.8555,
|
|
"mean_token_accuracy": 0.11735839322209358,
|
|
"num_tokens": 1695476.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.090408611297607,
|
|
"epoch": 0.8079071766222604,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.7525,
|
|
"mean_token_accuracy": 0.12118161767721176,
|
|
"num_tokens": 1704907.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.016019344329834,
|
|
"epoch": 0.812204555221315,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.7924,
|
|
"mean_token_accuracy": 0.12617168575525284,
|
|
"num_tokens": 1714564.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.132166576385498,
|
|
"epoch": 0.8165019338203696,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.8135,
|
|
"mean_token_accuracy": 0.12022659555077553,
|
|
"num_tokens": 1725285.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.00044469833374,
|
|
"epoch": 0.8207993124194242,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.8177,
|
|
"mean_token_accuracy": 0.12241263464093208,
|
|
"num_tokens": 1734331.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.126689529418945,
|
|
"epoch": 0.8250966910184787,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.749,
|
|
"mean_token_accuracy": 0.11530287116765976,
|
|
"num_tokens": 1742340.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.05500750541687,
|
|
"epoch": 0.8293940696175333,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.7383,
|
|
"mean_token_accuracy": 0.12545244619250298,
|
|
"num_tokens": 1751725.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.894489717483521,
|
|
"epoch": 0.8336914482165879,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.6736,
|
|
"mean_token_accuracy": 0.12856126353144645,
|
|
"num_tokens": 1760294.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.036704349517822,
|
|
"epoch": 0.8379888268156425,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.7265,
|
|
"mean_token_accuracy": 0.1231304183602333,
|
|
"num_tokens": 1768912.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.092654848098755,
|
|
"epoch": 0.842286205414697,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.9187,
|
|
"mean_token_accuracy": 0.12804483920335769,
|
|
"num_tokens": 1778633.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.090839195251465,
|
|
"epoch": 0.8465835840137517,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.7883,
|
|
"mean_token_accuracy": 0.12408955544233322,
|
|
"num_tokens": 1787275.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.0695414543151855,
|
|
"epoch": 0.8508809626128062,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.7844,
|
|
"mean_token_accuracy": 0.12348324134945869,
|
|
"num_tokens": 1795994.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.964667177200317,
|
|
"epoch": 0.8551783412118608,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.7175,
|
|
"mean_token_accuracy": 0.12602235972881318,
|
|
"num_tokens": 1806379.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.061655473709107,
|
|
"epoch": 0.8594757198109153,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.7479,
|
|
"mean_token_accuracy": 0.13024335727095604,
|
|
"num_tokens": 1816135.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8594757198109153,
|
|
"eval_entropy": 6.75515693050247,
|
|
"eval_loss": 6.752710819244385,
|
|
"eval_mean_token_accuracy": 0.12811107195175445,
|
|
"eval_num_tokens": 1816135.0,
|
|
"eval_runtime": 2.0604,
|
|
"eval_samples_per_second": 1722.442,
|
|
"eval_steps_per_second": 215.487,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.9897054672241214,
|
|
"epoch": 0.86377309840997,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004999998427807679,
|
|
"loss": 6.7314,
|
|
"mean_token_accuracy": 0.12282020673155784,
|
|
"num_tokens": 1824777.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.925821113586426,
|
|
"epoch": 0.8680704770090245,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004999992040780138,
|
|
"loss": 6.8085,
|
|
"mean_token_accuracy": 0.1247783549129963,
|
|
"num_tokens": 1833807.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.123036670684814,
|
|
"epoch": 0.8723678556080791,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999980740669294,
|
|
"loss": 6.754,
|
|
"mean_token_accuracy": 0.12499897480010987,
|
|
"num_tokens": 1843375.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.027141857147217,
|
|
"epoch": 0.8766652342071336,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999964527499823,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.12067028507590294,
|
|
"num_tokens": 1853036.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.018357038497925,
|
|
"epoch": 0.8809626128061883,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999943401307127,
|
|
"loss": 6.7605,
|
|
"mean_token_accuracy": 0.12497071847319603,
|
|
"num_tokens": 1862041.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.984006929397583,
|
|
"epoch": 0.8852599914052428,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999917362137337,
|
|
"loss": 6.6885,
|
|
"mean_token_accuracy": 0.12735832259058952,
|
|
"num_tokens": 1870707.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.964999151229859,
|
|
"epoch": 0.8895573700042974,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999886410047312,
|
|
"loss": 6.6849,
|
|
"mean_token_accuracy": 0.12543184384703637,
|
|
"num_tokens": 1879787.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.046022748947143,
|
|
"epoch": 0.8938547486033519,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999850545104638,
|
|
"loss": 6.7336,
|
|
"mean_token_accuracy": 0.12585699930787086,
|
|
"num_tokens": 1889413.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.9450146675109865,
|
|
"epoch": 0.8981521272024066,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999809767387633,
|
|
"loss": 6.7291,
|
|
"mean_token_accuracy": 0.12462790235877037,
|
|
"num_tokens": 1898283.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.982704973220825,
|
|
"epoch": 0.9024495058014611,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999764076985337,
|
|
"loss": 6.7474,
|
|
"mean_token_accuracy": 0.12953734770417213,
|
|
"num_tokens": 1907175.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.947793340682983,
|
|
"epoch": 0.9067468844005157,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999713473997519,
|
|
"loss": 6.7933,
|
|
"mean_token_accuracy": 0.12337937280535698,
|
|
"num_tokens": 1918223.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.053569555282593,
|
|
"epoch": 0.9110442629995703,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999657958534677,
|
|
"loss": 6.7435,
|
|
"mean_token_accuracy": 0.11936211958527565,
|
|
"num_tokens": 1928801.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.874362564086914,
|
|
"epoch": 0.9153416415986249,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999597530718034,
|
|
"loss": 6.7076,
|
|
"mean_token_accuracy": 0.12535862401127815,
|
|
"num_tokens": 1937406.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.924251508712769,
|
|
"epoch": 0.9196390201976794,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499953219067954,
|
|
"loss": 6.7025,
|
|
"mean_token_accuracy": 0.12463184967637062,
|
|
"num_tokens": 1947184.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.056308698654175,
|
|
"epoch": 0.923936398796734,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999461938561873,
|
|
"loss": 6.7241,
|
|
"mean_token_accuracy": 0.12476856112480164,
|
|
"num_tokens": 1956293.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.90220274925232,
|
|
"epoch": 0.9282337773957886,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999386774518432,
|
|
"loss": 6.6968,
|
|
"mean_token_accuracy": 0.12625648751854895,
|
|
"num_tokens": 1964791.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.965981435775757,
|
|
"epoch": 0.9325311559948432,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999306698713349,
|
|
"loss": 6.616,
|
|
"mean_token_accuracy": 0.12837354317307473,
|
|
"num_tokens": 1973754.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.929974555969238,
|
|
"epoch": 0.9368285345938977,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999221711321477,
|
|
"loss": 6.6857,
|
|
"mean_token_accuracy": 0.12695353776216506,
|
|
"num_tokens": 1983035.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.804391956329345,
|
|
"epoch": 0.9411259131929522,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999131812528393,
|
|
"loss": 6.7126,
|
|
"mean_token_accuracy": 0.12742481231689454,
|
|
"num_tokens": 1992584.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.0129533290863035,
|
|
"epoch": 0.9454232917920069,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00049990370025304,
|
|
"loss": 6.745,
|
|
"mean_token_accuracy": 0.1250165306031704,
|
|
"num_tokens": 2001876.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.9361108303070065,
|
|
"epoch": 0.9497206703910615,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998937281534526,
|
|
"loss": 6.6354,
|
|
"mean_token_accuracy": 0.1352070689201355,
|
|
"num_tokens": 2011067.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.00281867980957,
|
|
"epoch": 0.954018048990116,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998832649758521,
|
|
"loss": 6.7191,
|
|
"mean_token_accuracy": 0.12910578772425652,
|
|
"num_tokens": 2020763.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.846075534820557,
|
|
"epoch": 0.9583154275891707,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004998723107430862,
|
|
"loss": 6.702,
|
|
"mean_token_accuracy": 0.12597106099128724,
|
|
"num_tokens": 2029534.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.979312801361084,
|
|
"epoch": 0.9626128061882252,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998608654790741,
|
|
"loss": 6.6576,
|
|
"mean_token_accuracy": 0.12685178518295287,
|
|
"num_tokens": 2039143.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.840395832061768,
|
|
"epoch": 0.9669101847872797,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499848929208808,
|
|
"loss": 6.619,
|
|
"mean_token_accuracy": 0.13090287074446677,
|
|
"num_tokens": 2048253.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.833210182189942,
|
|
"epoch": 0.9712075633863343,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004998365019583519,
|
|
"loss": 6.6747,
|
|
"mean_token_accuracy": 0.13630941957235337,
|
|
"num_tokens": 2057234.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.008919525146484,
|
|
"epoch": 0.975504941985389,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998235837548417,
|
|
"loss": 6.7058,
|
|
"mean_token_accuracy": 0.12927891165018082,
|
|
"num_tokens": 2065431.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.887974071502685,
|
|
"epoch": 0.9798023205844435,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499810174626486,
|
|
"loss": 6.7146,
|
|
"mean_token_accuracy": 0.1267981804907322,
|
|
"num_tokens": 2074723.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.909135150909424,
|
|
"epoch": 0.984099699183498,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997962746025646,
|
|
"loss": 6.5835,
|
|
"mean_token_accuracy": 0.13582983165979384,
|
|
"num_tokens": 2084509.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.8790112972259525,
|
|
"epoch": 0.9883970777825526,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997818837134298,
|
|
"loss": 6.7192,
|
|
"mean_token_accuracy": 0.13046733066439628,
|
|
"num_tokens": 2093110.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.820547676086425,
|
|
"epoch": 0.9926944563816072,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997670019905057,
|
|
"loss": 6.5939,
|
|
"mean_token_accuracy": 0.12773325443267822,
|
|
"num_tokens": 2102355.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.849571800231933,
|
|
"epoch": 0.9969918349806618,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997516294662876,
|
|
"loss": 6.6207,
|
|
"mean_token_accuracy": 0.1278907351195812,
|
|
"num_tokens": 2110418.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.932281441158718,
|
|
"epoch": 1.0008594757198108,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997357661743433,
|
|
"loss": 6.6076,
|
|
"mean_token_accuracy": 0.13429299659199184,
|
|
"num_tokens": 2117866.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.776707983016967,
|
|
"epoch": 1.0051568543188656,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997194121493118,
|
|
"loss": 6.4353,
|
|
"mean_token_accuracy": 0.14019777849316598,
|
|
"num_tokens": 2126082.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.887734413146973,
|
|
"epoch": 1.0094542329179201,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997025674269037,
|
|
"loss": 6.4211,
|
|
"mean_token_accuracy": 0.13955733701586723,
|
|
"num_tokens": 2134042.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.774314117431641,
|
|
"epoch": 1.0137516115169747,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004996852320439013,
|
|
"loss": 6.4895,
|
|
"mean_token_accuracy": 0.13937605321407318,
|
|
"num_tokens": 2142570.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.8031017780303955,
|
|
"epoch": 1.0180489901160292,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996674060381578,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.13786159604787826,
|
|
"num_tokens": 2151310.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.884524583816528,
|
|
"epoch": 1.0223463687150838,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004996490894485985,
|
|
"loss": 6.4993,
|
|
"mean_token_accuracy": 0.1331610009074211,
|
|
"num_tokens": 2160662.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.801689147949219,
|
|
"epoch": 1.0266437473141383,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996302823152193,
|
|
"loss": 6.445,
|
|
"mean_token_accuracy": 0.13591438457369803,
|
|
"num_tokens": 2170067.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.76284008026123,
|
|
"epoch": 1.0309411259131929,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996109846790873,
|
|
"loss": 6.4084,
|
|
"mean_token_accuracy": 0.14033972024917601,
|
|
"num_tokens": 2178850.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.71863865852356,
|
|
"epoch": 1.0352385045122476,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995911965823412,
|
|
"loss": 6.4263,
|
|
"mean_token_accuracy": 0.1453915849328041,
|
|
"num_tokens": 2188307.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.847736549377442,
|
|
"epoch": 1.0395358831113022,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004995709180681899,
|
|
"loss": 6.4144,
|
|
"mean_token_accuracy": 0.1416982263326645,
|
|
"num_tokens": 2197026.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.729686546325683,
|
|
"epoch": 1.0438332617103567,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499550149180914,
|
|
"loss": 6.4003,
|
|
"mean_token_accuracy": 0.13990466818213462,
|
|
"num_tokens": 2205537.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.780020618438721,
|
|
"epoch": 1.0481306403094113,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995288899658641,
|
|
"loss": 6.4298,
|
|
"mean_token_accuracy": 0.1448238343000412,
|
|
"num_tokens": 2214508.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.842759847640991,
|
|
"epoch": 1.0524280189084658,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995071404694619,
|
|
"loss": 6.5391,
|
|
"mean_token_accuracy": 0.1354886084794998,
|
|
"num_tokens": 2223084.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.7924669742584225,
|
|
"epoch": 1.0567253975075204,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994849007391996,
|
|
"loss": 6.4679,
|
|
"mean_token_accuracy": 0.13138427063822747,
|
|
"num_tokens": 2231406.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.731750345230102,
|
|
"epoch": 1.061022776106575,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994621708236401,
|
|
"loss": 6.3805,
|
|
"mean_token_accuracy": 0.14119497835636138,
|
|
"num_tokens": 2239867.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.745153379440308,
|
|
"epoch": 1.0653201547056295,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000499438950772416,
|
|
"loss": 6.4467,
|
|
"mean_token_accuracy": 0.1372622825205326,
|
|
"num_tokens": 2248844.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.710582876205445,
|
|
"epoch": 1.0696175333046842,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994152406362311,
|
|
"loss": 6.3633,
|
|
"mean_token_accuracy": 0.14102791994810104,
|
|
"num_tokens": 2257599.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.773756074905395,
|
|
"epoch": 1.0739149119037388,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004993910404668586,
|
|
"loss": 6.418,
|
|
"mean_token_accuracy": 0.13638516888022423,
|
|
"num_tokens": 2266510.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.720381832122802,
|
|
"epoch": 1.0782122905027933,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499366350317142,
|
|
"loss": 6.4145,
|
|
"mean_token_accuracy": 0.1418795846402645,
|
|
"num_tokens": 2275462.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.712311601638794,
|
|
"epoch": 1.0825096691018479,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004993411702409948,
|
|
"loss": 6.3874,
|
|
"mean_token_accuracy": 0.1354715533554554,
|
|
"num_tokens": 2283826.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.76007399559021,
|
|
"epoch": 1.0868070477009024,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004993155002934002,
|
|
"loss": 6.3997,
|
|
"mean_token_accuracy": 0.13856483697891236,
|
|
"num_tokens": 2292967.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.8389280319213865,
|
|
"epoch": 1.091104426299957,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0004992893405304111,
|
|
"loss": 6.5262,
|
|
"mean_token_accuracy": 0.13781826868653296,
|
|
"num_tokens": 2302336.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.64991979598999,
|
|
"epoch": 1.0954018048990115,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00049926269100915,
|
|
"loss": 6.4293,
|
|
"mean_token_accuracy": 0.1432204395532608,
|
|
"num_tokens": 2311465.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.792691707611084,
|
|
"epoch": 1.0996991834980663,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992355517878087,
|
|
"loss": 6.542,
|
|
"mean_token_accuracy": 0.13071493357419967,
|
|
"num_tokens": 2320281.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.689556837081909,
|
|
"epoch": 1.1039965620971208,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004992079229256484,
|
|
"loss": 6.4431,
|
|
"mean_token_accuracy": 0.1360026031732559,
|
|
"num_tokens": 2329755.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.6757041931152346,
|
|
"epoch": 1.1082939406961754,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991798044829996,
|
|
"loss": 6.3861,
|
|
"mean_token_accuracy": 0.1369478650391102,
|
|
"num_tokens": 2338807.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.7733612060546875,
|
|
"epoch": 1.11259131929523,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004991511965212618,
|
|
"loss": 6.4719,
|
|
"mean_token_accuracy": 0.13780709579586983,
|
|
"num_tokens": 2348056.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.688971424102784,
|
|
"epoch": 1.1168886978942845,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991220991029032,
|
|
"loss": 6.4868,
|
|
"mean_token_accuracy": 0.13366840407252312,
|
|
"num_tokens": 2357780.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.773650407791138,
|
|
"epoch": 1.121186076493339,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000499092512291461,
|
|
"loss": 6.4446,
|
|
"mean_token_accuracy": 0.13651487827301026,
|
|
"num_tokens": 2367060.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.7718230247497555,
|
|
"epoch": 1.1254834550923936,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499062436151541,
|
|
"loss": 6.441,
|
|
"mean_token_accuracy": 0.1382215812802315,
|
|
"num_tokens": 2375751.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.800968360900879,
|
|
"epoch": 1.129780833691448,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004990318707488173,
|
|
"loss": 6.5069,
|
|
"mean_token_accuracy": 0.13017478883266448,
|
|
"num_tokens": 2385013.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.692961692810059,
|
|
"epoch": 1.1340782122905029,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004990008161500327,
|
|
"loss": 6.3937,
|
|
"mean_token_accuracy": 0.14006393477320672,
|
|
"num_tokens": 2392935.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.706206512451172,
|
|
"epoch": 1.1383755908895574,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000498969272422998,
|
|
"loss": 6.4188,
|
|
"mean_token_accuracy": 0.1468452200293541,
|
|
"num_tokens": 2401560.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.711210012435913,
|
|
"epoch": 1.142672969488612,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989372396365921,
|
|
"loss": 6.3447,
|
|
"mean_token_accuracy": 0.1455326870083809,
|
|
"num_tokens": 2410050.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.756243276596069,
|
|
"epoch": 1.1469703480876665,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989047178607618,
|
|
"loss": 6.4505,
|
|
"mean_token_accuracy": 0.13842038065195084,
|
|
"num_tokens": 2418980.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.671654081344604,
|
|
"epoch": 1.151267726686721,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004988717071665215,
|
|
"loss": 6.4407,
|
|
"mean_token_accuracy": 0.13684784546494483,
|
|
"num_tokens": 2427992.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.762688112258911,
|
|
"epoch": 1.1555651052857756,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004988382076259537,
|
|
"loss": 6.3572,
|
|
"mean_token_accuracy": 0.14135119169950486,
|
|
"num_tokens": 2436368.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.5892657279968265,
|
|
"epoch": 1.1598624838848304,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004988042193122077,
|
|
"loss": 6.3456,
|
|
"mean_token_accuracy": 0.14492984861135483,
|
|
"num_tokens": 2445499.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.752876138687133,
|
|
"epoch": 1.164159862483885,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004987697422995005,
|
|
"loss": 6.3818,
|
|
"mean_token_accuracy": 0.13490121066570282,
|
|
"num_tokens": 2454312.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.647862577438355,
|
|
"epoch": 1.1684572410829395,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004987347766631161,
|
|
"loss": 6.4437,
|
|
"mean_token_accuracy": 0.1407245770096779,
|
|
"num_tokens": 2462922.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.755164289474488,
|
|
"epoch": 1.172754619681994,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004986993224794055,
|
|
"loss": 6.4781,
|
|
"mean_token_accuracy": 0.13789629712700843,
|
|
"num_tokens": 2472195.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.6456316947937015,
|
|
"epoch": 1.1770519982810486,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004986633798257865,
|
|
"loss": 6.3829,
|
|
"mean_token_accuracy": 0.14376115351915358,
|
|
"num_tokens": 2481021.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.657115125656128,
|
|
"epoch": 1.181349376880103,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004986269487807434,
|
|
"loss": 6.405,
|
|
"mean_token_accuracy": 0.13883866667747496,
|
|
"num_tokens": 2490250.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.763047981262207,
|
|
"epoch": 1.1856467554791577,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000498590029423827,
|
|
"loss": 6.4581,
|
|
"mean_token_accuracy": 0.14272229447960855,
|
|
"num_tokens": 2499122.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.686977815628052,
|
|
"epoch": 1.1899441340782122,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004985526218356546,
|
|
"loss": 6.4227,
|
|
"mean_token_accuracy": 0.13726608753204345,
|
|
"num_tokens": 2508454.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.699887418746949,
|
|
"epoch": 1.1942415126772667,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004985147260979093,
|
|
"loss": 6.3632,
|
|
"mean_token_accuracy": 0.1465839110314846,
|
|
"num_tokens": 2517353.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.691904354095459,
|
|
"epoch": 1.1985388912763215,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004984763422933402,
|
|
"loss": 6.3821,
|
|
"mean_token_accuracy": 0.14337702393531798,
|
|
"num_tokens": 2526321.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.6859358787536625,
|
|
"epoch": 1.202836269875376,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004984374705057623,
|
|
"loss": 6.4144,
|
|
"mean_token_accuracy": 0.14242582842707635,
|
|
"num_tokens": 2535924.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.640392780303955,
|
|
"epoch": 1.2071336484744306,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983981108200561,
|
|
"loss": 6.3922,
|
|
"mean_token_accuracy": 0.1401688925921917,
|
|
"num_tokens": 2545606.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.649671459197998,
|
|
"epoch": 1.2114310270734852,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983582633221672,
|
|
"loss": 6.3859,
|
|
"mean_token_accuracy": 0.1407300591468811,
|
|
"num_tokens": 2554947.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.765527582168579,
|
|
"epoch": 1.2157284056725397,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004983179280991068,
|
|
"loss": 6.5354,
|
|
"mean_token_accuracy": 0.13627680763602257,
|
|
"num_tokens": 2564462.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.688222122192383,
|
|
"epoch": 1.2200257842715942,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004982771052389508,
|
|
"loss": 6.3743,
|
|
"mean_token_accuracy": 0.1444454774260521,
|
|
"num_tokens": 2573124.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.700618696212769,
|
|
"epoch": 1.224323162870649,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004982357948308401,
|
|
"loss": 6.4798,
|
|
"mean_token_accuracy": 0.13040754944086075,
|
|
"num_tokens": 2581829.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.7136975765228275,
|
|
"epoch": 1.2286205414697036,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004981939969649799,
|
|
"loss": 6.3405,
|
|
"mean_token_accuracy": 0.1422662131488323,
|
|
"num_tokens": 2590631.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.661464500427246,
|
|
"epoch": 1.232917920068758,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004981517117326404,
|
|
"loss": 6.4484,
|
|
"mean_token_accuracy": 0.13987314701080322,
|
|
"num_tokens": 2600684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.6479767799377445,
|
|
"epoch": 1.2372152986678127,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004981089392261553,
|
|
"loss": 6.3605,
|
|
"mean_token_accuracy": 0.14449947997927665,
|
|
"num_tokens": 2609667.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.643135976791382,
|
|
"epoch": 1.2415126772668672,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000498065679538923,
|
|
"loss": 6.4317,
|
|
"mean_token_accuracy": 0.14703501164913177,
|
|
"num_tokens": 2620025.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.672731685638428,
|
|
"epoch": 1.2458100558659218,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004980219327654049,
|
|
"loss": 6.351,
|
|
"mean_token_accuracy": 0.14008775800466539,
|
|
"num_tokens": 2629032.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.605780506134034,
|
|
"epoch": 1.2501074344649763,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000497977699001127,
|
|
"loss": 6.3357,
|
|
"mean_token_accuracy": 0.1428795799612999,
|
|
"num_tokens": 2638303.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.698618459701538,
|
|
"epoch": 1.2544048130640308,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004979329783426778,
|
|
"loss": 6.3527,
|
|
"mean_token_accuracy": 0.14518981352448462,
|
|
"num_tokens": 2647902.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.619544601440429,
|
|
"epoch": 1.2587021916630854,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978877708877094,
|
|
"loss": 6.4046,
|
|
"mean_token_accuracy": 0.1414396196603775,
|
|
"num_tokens": 2657902.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.67303991317749,
|
|
"epoch": 1.2629995702621402,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978420767349368,
|
|
"loss": 6.3504,
|
|
"mean_token_accuracy": 0.14340997561812402,
|
|
"num_tokens": 2667082.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.647952270507813,
|
|
"epoch": 1.2672969488611947,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977958959841379,
|
|
"loss": 6.4223,
|
|
"mean_token_accuracy": 0.1364084042608738,
|
|
"num_tokens": 2676855.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.6442427158355715,
|
|
"epoch": 1.2715943274602493,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497749228736153,
|
|
"loss": 6.3546,
|
|
"mean_token_accuracy": 0.145116026699543,
|
|
"num_tokens": 2685750.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.597840929031372,
|
|
"epoch": 1.2758917060593038,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004977020750928845,
|
|
"loss": 6.4075,
|
|
"mean_token_accuracy": 0.14761355221271516,
|
|
"num_tokens": 2695272.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.709882497787476,
|
|
"epoch": 1.2801890846583583,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004976544351572973,
|
|
"loss": 6.3504,
|
|
"mean_token_accuracy": 0.1418570265173912,
|
|
"num_tokens": 2704806.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.533363771438599,
|
|
"epoch": 1.2844864632574131,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004976063090334179,
|
|
"loss": 6.4036,
|
|
"mean_token_accuracy": 0.1452034071087837,
|
|
"num_tokens": 2713521.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.7042053699493405,
|
|
"epoch": 1.2887838418564677,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004975576968263346,
|
|
"loss": 6.3966,
|
|
"mean_token_accuracy": 0.1381194919347763,
|
|
"num_tokens": 2721848.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2887838418564677,
|
|
"eval_entropy": 6.494678375957249,
|
|
"eval_loss": 6.482933044433594,
|
|
"eval_mean_token_accuracy": 0.14236528785513328,
|
|
"eval_num_tokens": 2721848.0,
|
|
"eval_runtime": 2.0538,
|
|
"eval_samples_per_second": 1728.039,
|
|
"eval_steps_per_second": 216.187,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.592136430740356,
|
|
"epoch": 1.2930812204555222,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000497508598642197,
|
|
"loss": 6.3613,
|
|
"mean_token_accuracy": 0.14413030222058296,
|
|
"num_tokens": 2731473.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.610020494461059,
|
|
"epoch": 1.2973785990545768,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497459014588216,
|
|
"loss": 6.4326,
|
|
"mean_token_accuracy": 0.141157578676939,
|
|
"num_tokens": 2739867.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.684322929382324,
|
|
"epoch": 1.3016759776536313,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000497408944772663,
|
|
"loss": 6.3442,
|
|
"mean_token_accuracy": 0.14187844544649125,
|
|
"num_tokens": 2748903.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.512551116943359,
|
|
"epoch": 1.3059733562526858,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004973583893048707,
|
|
"loss": 6.3389,
|
|
"mean_token_accuracy": 0.14152248129248618,
|
|
"num_tokens": 2757711.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.74653639793396,
|
|
"epoch": 1.3102707348517404,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004973073482952321,
|
|
"loss": 6.358,
|
|
"mean_token_accuracy": 0.140853676199913,
|
|
"num_tokens": 2765633.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.572407674789429,
|
|
"epoch": 1.314568113450795,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004972558218552004,
|
|
"loss": 6.3982,
|
|
"mean_token_accuracy": 0.14053191468119622,
|
|
"num_tokens": 2774495.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.645643854141236,
|
|
"epoch": 1.3188654920498495,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004972038100972885,
|
|
"loss": 6.4066,
|
|
"mean_token_accuracy": 0.1426756389439106,
|
|
"num_tokens": 2782665.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.549836540222168,
|
|
"epoch": 1.323162870648904,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004971513131350697,
|
|
"loss": 6.356,
|
|
"mean_token_accuracy": 0.13861292153596877,
|
|
"num_tokens": 2791394.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.566079998016358,
|
|
"epoch": 1.3274602492479588,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004970983310831759,
|
|
"loss": 6.3437,
|
|
"mean_token_accuracy": 0.1422226123511791,
|
|
"num_tokens": 2800488.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.6656012535095215,
|
|
"epoch": 1.3317576278470133,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004970448640572989,
|
|
"loss": 6.4644,
|
|
"mean_token_accuracy": 0.14133307337760925,
|
|
"num_tokens": 2810116.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.59561824798584,
|
|
"epoch": 1.336055006446068,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004969909121741895,
|
|
"loss": 6.2592,
|
|
"mean_token_accuracy": 0.14750397205352783,
|
|
"num_tokens": 2819205.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.559555625915527,
|
|
"epoch": 1.3403523850451224,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004969364755516569,
|
|
"loss": 6.3311,
|
|
"mean_token_accuracy": 0.14398850798606871,
|
|
"num_tokens": 2828017.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.688138008117676,
|
|
"epoch": 1.344649763644177,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004968815543085689,
|
|
"loss": 6.3815,
|
|
"mean_token_accuracy": 0.145321074873209,
|
|
"num_tokens": 2837125.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.569426822662353,
|
|
"epoch": 1.3489471422432318,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004968261485648516,
|
|
"loss": 6.3921,
|
|
"mean_token_accuracy": 0.14212561994791031,
|
|
"num_tokens": 2845438.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.608628225326538,
|
|
"epoch": 1.3532445208422863,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000496770258441489,
|
|
"loss": 6.3689,
|
|
"mean_token_accuracy": 0.1471138596534729,
|
|
"num_tokens": 2854389.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.556783771514892,
|
|
"epoch": 1.3575418994413408,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004967138840605228,
|
|
"loss": 6.3281,
|
|
"mean_token_accuracy": 0.14712274819612503,
|
|
"num_tokens": 2863654.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.517911720275879,
|
|
"epoch": 1.3618392780403954,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000496657025545052,
|
|
"loss": 6.2482,
|
|
"mean_token_accuracy": 0.15075734853744507,
|
|
"num_tokens": 2872871.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.5070977210998535,
|
|
"epoch": 1.36613665663945,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000496599683019233,
|
|
"loss": 6.3373,
|
|
"mean_token_accuracy": 0.1449791297316551,
|
|
"num_tokens": 2881140.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.6506085872650145,
|
|
"epoch": 1.3704340352385045,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000496541856608279,
|
|
"loss": 6.3251,
|
|
"mean_token_accuracy": 0.14629032611846923,
|
|
"num_tokens": 2889945.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.464802026748657,
|
|
"epoch": 1.374731413837559,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004964835464384595,
|
|
"loss": 6.254,
|
|
"mean_token_accuracy": 0.14956037551164628,
|
|
"num_tokens": 2898897.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.606829452514648,
|
|
"epoch": 1.3790287924366136,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000496424752637101,
|
|
"loss": 6.2819,
|
|
"mean_token_accuracy": 0.15412394404411317,
|
|
"num_tokens": 2907717.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.513754224777221,
|
|
"epoch": 1.3833261710356681,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004963654753325853,
|
|
"loss": 6.2693,
|
|
"mean_token_accuracy": 0.1435668349266052,
|
|
"num_tokens": 2916213.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.6343999862670895,
|
|
"epoch": 1.387623549634723,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004963057146543505,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.13986597284674646,
|
|
"num_tokens": 2925706.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.570179843902588,
|
|
"epoch": 1.3919209282337774,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00049624547073289,
|
|
"loss": 6.3511,
|
|
"mean_token_accuracy": 0.13794696033000947,
|
|
"num_tokens": 2934464.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.570999479293823,
|
|
"epoch": 1.396218306832832,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004961847436997526,
|
|
"loss": 6.2482,
|
|
"mean_token_accuracy": 0.1511821575462818,
|
|
"num_tokens": 2944095.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.450803470611572,
|
|
"epoch": 1.4005156854318865,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004961235336875416,
|
|
"loss": 6.249,
|
|
"mean_token_accuracy": 0.1513315513730049,
|
|
"num_tokens": 2953357.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.5238546371459964,
|
|
"epoch": 1.404813064030941,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004960618408299154,
|
|
"loss": 6.4089,
|
|
"mean_token_accuracy": 0.1346985176205635,
|
|
"num_tokens": 2963020.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.61925859451294,
|
|
"epoch": 1.4091104426299956,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004959996652615865,
|
|
"loss": 6.2427,
|
|
"mean_token_accuracy": 0.1468616619706154,
|
|
"num_tokens": 2971955.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.584984397888183,
|
|
"epoch": 1.4134078212290504,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004959370071183216,
|
|
"loss": 6.3097,
|
|
"mean_token_accuracy": 0.14391712918877603,
|
|
"num_tokens": 2980662.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.6156212329864506,
|
|
"epoch": 1.417705199828105,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004958738665369407,
|
|
"loss": 6.439,
|
|
"mean_token_accuracy": 0.12904247269034386,
|
|
"num_tokens": 2990038.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.566392660140991,
|
|
"epoch": 1.4220025784271595,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004958102436553179,
|
|
"loss": 6.3627,
|
|
"mean_token_accuracy": 0.1401166081428528,
|
|
"num_tokens": 2999835.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.622867441177368,
|
|
"epoch": 1.426299957026214,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00049574613861238,
|
|
"loss": 6.3528,
|
|
"mean_token_accuracy": 0.1401872843503952,
|
|
"num_tokens": 3009593.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.564433908462524,
|
|
"epoch": 1.4305973356252686,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004956815515481069,
|
|
"loss": 6.3748,
|
|
"mean_token_accuracy": 0.14576212018728257,
|
|
"num_tokens": 3019187.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.528054189682007,
|
|
"epoch": 1.4348947142243231,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004956164826035309,
|
|
"loss": 6.2839,
|
|
"mean_token_accuracy": 0.14402172416448594,
|
|
"num_tokens": 3027875.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.481614637374878,
|
|
"epoch": 1.4391920928233777,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004955509319207363,
|
|
"loss": 6.3184,
|
|
"mean_token_accuracy": 0.14420104324817656,
|
|
"num_tokens": 3036902.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.46042537689209,
|
|
"epoch": 1.4434894714224322,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004954848996428601,
|
|
"loss": 6.2969,
|
|
"mean_token_accuracy": 0.1498032405972481,
|
|
"num_tokens": 3046653.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.64046082496643,
|
|
"epoch": 1.4477868500214868,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00049541838591409,
|
|
"loss": 6.3977,
|
|
"mean_token_accuracy": 0.14052897915244103,
|
|
"num_tokens": 3056273.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.529829502105713,
|
|
"epoch": 1.4520842286205415,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004953513908796657,
|
|
"loss": 6.2999,
|
|
"mean_token_accuracy": 0.13732842430472375,
|
|
"num_tokens": 3065662.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.594562721252442,
|
|
"epoch": 1.456381607219596,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004952839146858773,
|
|
"loss": 6.3277,
|
|
"mean_token_accuracy": 0.14757051467895507,
|
|
"num_tokens": 3073970.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.531829500198365,
|
|
"epoch": 1.4606789858186506,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004952159574800658,
|
|
"loss": 6.3209,
|
|
"mean_token_accuracy": 0.14381522089242935,
|
|
"num_tokens": 3082500.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.566446447372437,
|
|
"epoch": 1.4649763644177052,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004951475194106229,
|
|
"loss": 6.2777,
|
|
"mean_token_accuracy": 0.14633866250514985,
|
|
"num_tokens": 3091574.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.512380361557007,
|
|
"epoch": 1.4692737430167597,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004950786006269898,
|
|
"loss": 6.3852,
|
|
"mean_token_accuracy": 0.13938545510172845,
|
|
"num_tokens": 3102402.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.59727463722229,
|
|
"epoch": 1.4735711216158143,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004950092012796576,
|
|
"loss": 6.2072,
|
|
"mean_token_accuracy": 0.15373199433088303,
|
|
"num_tokens": 3111347.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.486224889755249,
|
|
"epoch": 1.477868500214869,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004949393215201666,
|
|
"loss": 6.2833,
|
|
"mean_token_accuracy": 0.14614666104316712,
|
|
"num_tokens": 3120018.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.4936051845550535,
|
|
"epoch": 1.4821658788139236,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004948689615011065,
|
|
"loss": 6.3484,
|
|
"mean_token_accuracy": 0.13831731379032136,
|
|
"num_tokens": 3129669.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.6139086246490475,
|
|
"epoch": 1.4864632574129781,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004947981213761154,
|
|
"loss": 6.2794,
|
|
"mean_token_accuracy": 0.15020036697387695,
|
|
"num_tokens": 3139112.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.510036754608154,
|
|
"epoch": 1.4907606360120327,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004947268012998797,
|
|
"loss": 6.2427,
|
|
"mean_token_accuracy": 0.15479698032140732,
|
|
"num_tokens": 3148437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.490271472930909,
|
|
"epoch": 1.4950580146110872,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000494655001428134,
|
|
"loss": 6.2146,
|
|
"mean_token_accuracy": 0.15289759933948516,
|
|
"num_tokens": 3158165.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.521289396286011,
|
|
"epoch": 1.4993553932101418,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004945827219176604,
|
|
"loss": 6.3026,
|
|
"mean_token_accuracy": 0.1522263005375862,
|
|
"num_tokens": 3167262.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.448360395431519,
|
|
"epoch": 1.5036527718091963,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004945099629262888,
|
|
"loss": 6.2841,
|
|
"mean_token_accuracy": 0.14779476150870324,
|
|
"num_tokens": 3176696.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.6200721740722654,
|
|
"epoch": 1.5079501504082509,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004944367246128954,
|
|
"loss": 6.3626,
|
|
"mean_token_accuracy": 0.1411810874938965,
|
|
"num_tokens": 3185857.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.497649145126343,
|
|
"epoch": 1.5122475290073054,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004943630071374036,
|
|
"loss": 6.2129,
|
|
"mean_token_accuracy": 0.15686369836330413,
|
|
"num_tokens": 3194687.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.447890901565552,
|
|
"epoch": 1.51654490760636,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004942888106607828,
|
|
"loss": 6.2715,
|
|
"mean_token_accuracy": 0.14421172440052032,
|
|
"num_tokens": 3204913.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.556134462356567,
|
|
"epoch": 1.5208422862054147,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004942141353450486,
|
|
"loss": 6.2587,
|
|
"mean_token_accuracy": 0.14712465703487396,
|
|
"num_tokens": 3213312.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.4831544876098635,
|
|
"epoch": 1.5251396648044693,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004941389813532619,
|
|
"loss": 6.1822,
|
|
"mean_token_accuracy": 0.1586100459098816,
|
|
"num_tokens": 3222992.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.385056638717652,
|
|
"epoch": 1.5294370434035238,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000494063348849529,
|
|
"loss": 6.2213,
|
|
"mean_token_accuracy": 0.15424711555242537,
|
|
"num_tokens": 3232836.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.574507141113282,
|
|
"epoch": 1.5337344220025786,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004939872379990011,
|
|
"loss": 6.3769,
|
|
"mean_token_accuracy": 0.14118290394544603,
|
|
"num_tokens": 3243171.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.56547212600708,
|
|
"epoch": 1.5380318006016331,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004939106489678739,
|
|
"loss": 6.2954,
|
|
"mean_token_accuracy": 0.15190573930740356,
|
|
"num_tokens": 3251995.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.440187692642212,
|
|
"epoch": 1.5423291792006877,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000493833581923387,
|
|
"loss": 6.2474,
|
|
"mean_token_accuracy": 0.14897289276123046,
|
|
"num_tokens": 3260841.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.5475788593292235,
|
|
"epoch": 1.5466265577997422,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004937560370338244,
|
|
"loss": 6.382,
|
|
"mean_token_accuracy": 0.14083073958754538,
|
|
"num_tokens": 3270979.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.536606645584106,
|
|
"epoch": 1.5509239363987968,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000493678014468513,
|
|
"loss": 6.307,
|
|
"mean_token_accuracy": 0.1528750814497471,
|
|
"num_tokens": 3279848.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.46652889251709,
|
|
"epoch": 1.5552213149978513,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004935995143978227,
|
|
"loss": 6.311,
|
|
"mean_token_accuracy": 0.14874453395605086,
|
|
"num_tokens": 3289172.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.480955171585083,
|
|
"epoch": 1.5595186935969059,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004935205369931664,
|
|
"loss": 6.2107,
|
|
"mean_token_accuracy": 0.15236888080835342,
|
|
"num_tokens": 3297432.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.62280158996582,
|
|
"epoch": 1.5638160721959604,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004934410824269992,
|
|
"loss": 6.2391,
|
|
"mean_token_accuracy": 0.1460478588938713,
|
|
"num_tokens": 3307486.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.396580219268799,
|
|
"epoch": 1.568113450795015,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004933611508728182,
|
|
"loss": 6.2234,
|
|
"mean_token_accuracy": 0.15543457493185997,
|
|
"num_tokens": 3316296.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.48117151260376,
|
|
"epoch": 1.5724108293940695,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000493280742505162,
|
|
"loss": 6.2496,
|
|
"mean_token_accuracy": 0.14565204530954362,
|
|
"num_tokens": 3326080.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.399107646942139,
|
|
"epoch": 1.576708207993124,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004931998574996102,
|
|
"loss": 6.1637,
|
|
"mean_token_accuracy": 0.1557439833879471,
|
|
"num_tokens": 3334826.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.395985460281372,
|
|
"epoch": 1.5810055865921788,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004931184960327832,
|
|
"loss": 6.166,
|
|
"mean_token_accuracy": 0.15891503393650055,
|
|
"num_tokens": 3343261.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.439464569091797,
|
|
"epoch": 1.5853029651912334,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0004930366582823421,
|
|
"loss": 6.2095,
|
|
"mean_token_accuracy": 0.14784578159451484,
|
|
"num_tokens": 3352513.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.446910238265991,
|
|
"epoch": 1.589600343790288,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004929543444269879,
|
|
"loss": 6.2679,
|
|
"mean_token_accuracy": 0.15295199751853944,
|
|
"num_tokens": 3361577.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.4689103126525875,
|
|
"epoch": 1.5938977223893425,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000492871554646461,
|
|
"loss": 6.327,
|
|
"mean_token_accuracy": 0.14370332658290863,
|
|
"num_tokens": 3370591.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.443254470825195,
|
|
"epoch": 1.5981951009883972,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004927882891215413,
|
|
"loss": 6.2437,
|
|
"mean_token_accuracy": 0.14615294709801674,
|
|
"num_tokens": 3379761.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.549100685119629,
|
|
"epoch": 1.6024924795874518,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004927045480340475,
|
|
"loss": 6.3212,
|
|
"mean_token_accuracy": 0.1414845257997513,
|
|
"num_tokens": 3388974.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.428477334976196,
|
|
"epoch": 1.6067898581865063,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004926203315668363,
|
|
"loss": 6.2385,
|
|
"mean_token_accuracy": 0.15081687197089194,
|
|
"num_tokens": 3398339.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.499061059951782,
|
|
"epoch": 1.6110872367855609,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004925356399038032,
|
|
"loss": 6.2121,
|
|
"mean_token_accuracy": 0.15119217038154603,
|
|
"num_tokens": 3408292.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.460348415374756,
|
|
"epoch": 1.6153846153846154,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004924504732298808,
|
|
"loss": 6.1809,
|
|
"mean_token_accuracy": 0.15673429295420646,
|
|
"num_tokens": 3417057.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.498525190353393,
|
|
"epoch": 1.61968199398367,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004923648317310391,
|
|
"loss": 6.2886,
|
|
"mean_token_accuracy": 0.15057691931724548,
|
|
"num_tokens": 3425830.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.466361808776855,
|
|
"epoch": 1.6239793725827245,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004922787155942849,
|
|
"loss": 6.3261,
|
|
"mean_token_accuracy": 0.14087508171796798,
|
|
"num_tokens": 3435513.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.480417251586914,
|
|
"epoch": 1.628276751181779,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004921921250076611,
|
|
"loss": 6.2319,
|
|
"mean_token_accuracy": 0.1488749422132969,
|
|
"num_tokens": 3444684.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.398703765869141,
|
|
"epoch": 1.6325741297808336,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004921050601602475,
|
|
"loss": 6.309,
|
|
"mean_token_accuracy": 0.15032647401094437,
|
|
"num_tokens": 3453454.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.512422227859497,
|
|
"epoch": 1.6368715083798882,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004920175212421587,
|
|
"loss": 6.2317,
|
|
"mean_token_accuracy": 0.1462756022810936,
|
|
"num_tokens": 3463228.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.298534250259399,
|
|
"epoch": 1.6411688869789427,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004919295084445445,
|
|
"loss": 6.1203,
|
|
"mean_token_accuracy": 0.15622290521860122,
|
|
"num_tokens": 3472131.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.46199779510498,
|
|
"epoch": 1.6454662655779975,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004918410219595899,
|
|
"loss": 6.1947,
|
|
"mean_token_accuracy": 0.15805622637271882,
|
|
"num_tokens": 3480642.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.536061143875122,
|
|
"epoch": 1.649763644177052,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000491752061980514,
|
|
"loss": 6.1748,
|
|
"mean_token_accuracy": 0.15212914645671843,
|
|
"num_tokens": 3489346.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.385542201995849,
|
|
"epoch": 1.6540610227761066,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004916626287015697,
|
|
"loss": 6.2236,
|
|
"mean_token_accuracy": 0.1506744407117367,
|
|
"num_tokens": 3498473.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.4339292526245115,
|
|
"epoch": 1.658358401375161,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004915727223180436,
|
|
"loss": 6.2184,
|
|
"mean_token_accuracy": 0.1503354400396347,
|
|
"num_tokens": 3507415.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.472232723236084,
|
|
"epoch": 1.6626557799742159,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004914823430262554,
|
|
"loss": 6.3466,
|
|
"mean_token_accuracy": 0.13937689363956451,
|
|
"num_tokens": 3516873.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.475211191177368,
|
|
"epoch": 1.6669531585732704,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004913914910235573,
|
|
"loss": 6.2023,
|
|
"mean_token_accuracy": 0.15309734791517257,
|
|
"num_tokens": 3525047.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.334531784057617,
|
|
"epoch": 1.671250537172325,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004913001665083337,
|
|
"loss": 6.2098,
|
|
"mean_token_accuracy": 0.1510941930115223,
|
|
"num_tokens": 3534354.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.499793291091919,
|
|
"epoch": 1.6755479157713795,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004912083696800008,
|
|
"loss": 6.2384,
|
|
"mean_token_accuracy": 0.14515842348337174,
|
|
"num_tokens": 3543830.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.334777593612671,
|
|
"epoch": 1.679845294370434,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004911161007390063,
|
|
"loss": 6.1344,
|
|
"mean_token_accuracy": 0.1552545964717865,
|
|
"num_tokens": 3552314.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.398986530303955,
|
|
"epoch": 1.6841426729694886,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004910233598868287,
|
|
"loss": 6.2232,
|
|
"mean_token_accuracy": 0.14675267040729523,
|
|
"num_tokens": 3561656.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.426092958450317,
|
|
"epoch": 1.6884400515685432,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004909301473259769,
|
|
"loss": 6.2232,
|
|
"mean_token_accuracy": 0.14848204478621482,
|
|
"num_tokens": 3571784.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.454012489318847,
|
|
"epoch": 1.6927374301675977,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004908364632599899,
|
|
"loss": 6.1775,
|
|
"mean_token_accuracy": 0.15773458033800125,
|
|
"num_tokens": 3580626.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.337477779388427,
|
|
"epoch": 1.6970348087666522,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004907423078934362,
|
|
"loss": 6.2001,
|
|
"mean_token_accuracy": 0.14792972654104233,
|
|
"num_tokens": 3589916.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.395978736877441,
|
|
"epoch": 1.7013321873657068,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004906476814319134,
|
|
"loss": 6.2045,
|
|
"mean_token_accuracy": 0.15436216294765473,
|
|
"num_tokens": 3599128.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.384798145294189,
|
|
"epoch": 1.7056295659647613,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004905525840820481,
|
|
"loss": 6.2156,
|
|
"mean_token_accuracy": 0.1487440824508667,
|
|
"num_tokens": 3608764.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.519760847091675,
|
|
"epoch": 1.709926944563816,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004904570160514948,
|
|
"loss": 6.2587,
|
|
"mean_token_accuracy": 0.14064486026763917,
|
|
"num_tokens": 3619082.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.396596527099609,
|
|
"epoch": 1.7142243231628707,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004903609775489358,
|
|
"loss": 6.2232,
|
|
"mean_token_accuracy": 0.14829822033643722,
|
|
"num_tokens": 3628695.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.453386020660401,
|
|
"epoch": 1.7185217017619252,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004902644687840809,
|
|
"loss": 6.2106,
|
|
"mean_token_accuracy": 0.14628567397594452,
|
|
"num_tokens": 3637599.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7185217017619252,
|
|
"eval_entropy": 6.120181280213433,
|
|
"eval_loss": 6.287801742553711,
|
|
"eval_mean_token_accuracy": 0.15146609128931085,
|
|
"eval_num_tokens": 3637599.0,
|
|
"eval_runtime": 2.0623,
|
|
"eval_samples_per_second": 1720.853,
|
|
"eval_steps_per_second": 215.289,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.389330768585205,
|
|
"epoch": 1.7228190803609797,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004901674899676667,
|
|
"loss": 6.189,
|
|
"mean_token_accuracy": 0.15087567865848542,
|
|
"num_tokens": 3647406.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.32288761138916,
|
|
"epoch": 1.7271164589600345,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004900700413114561,
|
|
"loss": 6.0845,
|
|
"mean_token_accuracy": 0.15229684859514236,
|
|
"num_tokens": 3656531.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.2982823848724365,
|
|
"epoch": 1.731413837559089,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000489972123028238,
|
|
"loss": 6.1711,
|
|
"mean_token_accuracy": 0.14639344438910484,
|
|
"num_tokens": 3664922.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.42927360534668,
|
|
"epoch": 1.7357112161581436,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004898737353318268,
|
|
"loss": 6.114,
|
|
"mean_token_accuracy": 0.15603691339492798,
|
|
"num_tokens": 3673283.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.379903554916382,
|
|
"epoch": 1.7400085947571982,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000489774878437062,
|
|
"loss": 6.2432,
|
|
"mean_token_accuracy": 0.1512456476688385,
|
|
"num_tokens": 3681760.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.362637662887574,
|
|
"epoch": 1.7443059733562527,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004896755525598074,
|
|
"loss": 6.0576,
|
|
"mean_token_accuracy": 0.15525488257408143,
|
|
"num_tokens": 3689408.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.350458097457886,
|
|
"epoch": 1.7486033519553073,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004895757579169511,
|
|
"loss": 6.1868,
|
|
"mean_token_accuracy": 0.1519346058368683,
|
|
"num_tokens": 3697904.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.549949407577515,
|
|
"epoch": 1.7529007305543618,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004894754947264047,
|
|
"loss": 6.2025,
|
|
"mean_token_accuracy": 0.1540897861123085,
|
|
"num_tokens": 3706704.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.33614501953125,
|
|
"epoch": 1.7571981091534163,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000489374763207103,
|
|
"loss": 6.2858,
|
|
"mean_token_accuracy": 0.14759851694107057,
|
|
"num_tokens": 3715690.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.4482136249542235,
|
|
"epoch": 1.761495487752471,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004892735635790033,
|
|
"loss": 6.0651,
|
|
"mean_token_accuracy": 0.16219264268875122,
|
|
"num_tokens": 3724835.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.303025627136231,
|
|
"epoch": 1.7657928663515254,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000489171896063085,
|
|
"loss": 6.0978,
|
|
"mean_token_accuracy": 0.1608540341258049,
|
|
"num_tokens": 3733977.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.440810489654541,
|
|
"epoch": 1.77009024495058,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004890697608813495,
|
|
"loss": 6.2166,
|
|
"mean_token_accuracy": 0.14737534075975417,
|
|
"num_tokens": 3742665.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.50860743522644,
|
|
"epoch": 1.7743876235496348,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004889671582568193,
|
|
"loss": 6.2866,
|
|
"mean_token_accuracy": 0.15046041160821916,
|
|
"num_tokens": 3751647.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.323904037475586,
|
|
"epoch": 1.7786850021486893,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004888640884135374,
|
|
"loss": 6.1804,
|
|
"mean_token_accuracy": 0.14905625879764556,
|
|
"num_tokens": 3760852.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.3692279815673825,
|
|
"epoch": 1.7829823807477438,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004887605515765671,
|
|
"loss": 6.146,
|
|
"mean_token_accuracy": 0.1545763321220875,
|
|
"num_tokens": 3768640.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.432651662826538,
|
|
"epoch": 1.7872797593467986,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004886565479719914,
|
|
"loss": 6.1701,
|
|
"mean_token_accuracy": 0.1504896029829979,
|
|
"num_tokens": 3776504.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.4639040470123295,
|
|
"epoch": 1.7915771379458532,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004885520778269128,
|
|
"loss": 6.1968,
|
|
"mean_token_accuracy": 0.15468488037586212,
|
|
"num_tokens": 3786353.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.380429744720459,
|
|
"epoch": 1.7958745165449077,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004884471413694523,
|
|
"loss": 6.2326,
|
|
"mean_token_accuracy": 0.14940588921308517,
|
|
"num_tokens": 3795902.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.3466850280761715,
|
|
"epoch": 1.8001718951439623,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004883417388287491,
|
|
"loss": 6.1431,
|
|
"mean_token_accuracy": 0.14718958735466003,
|
|
"num_tokens": 3805986.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.3597740650177,
|
|
"epoch": 1.8044692737430168,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004882358704349603,
|
|
"loss": 6.2747,
|
|
"mean_token_accuracy": 0.15220490992069244,
|
|
"num_tokens": 3814915.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.366986703872681,
|
|
"epoch": 1.8087666523420713,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004881295364192601,
|
|
"loss": 6.1506,
|
|
"mean_token_accuracy": 0.15957469791173934,
|
|
"num_tokens": 3823966.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.475821685791016,
|
|
"epoch": 1.813064030941126,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004880227370138394,
|
|
"loss": 6.212,
|
|
"mean_token_accuracy": 0.14951324909925462,
|
|
"num_tokens": 3832775.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.301672267913818,
|
|
"epoch": 1.8173614095401804,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004879154724519057,
|
|
"loss": 6.1316,
|
|
"mean_token_accuracy": 0.15576981902122497,
|
|
"num_tokens": 3842808.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.454287385940551,
|
|
"epoch": 1.821658788139235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004878077429676816,
|
|
"loss": 6.2649,
|
|
"mean_token_accuracy": 0.14898920953273773,
|
|
"num_tokens": 3853303.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.3901642799377445,
|
|
"epoch": 1.8259561667382895,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004876995487964054,
|
|
"loss": 6.1853,
|
|
"mean_token_accuracy": 0.14685731381177902,
|
|
"num_tokens": 3862462.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.411391401290894,
|
|
"epoch": 1.830253545337344,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048759089017432996,
|
|
"loss": 6.293,
|
|
"mean_token_accuracy": 0.14782755076885223,
|
|
"num_tokens": 3871596.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.432213401794433,
|
|
"epoch": 1.8345509239363988,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004874817673387222,
|
|
"loss": 6.1972,
|
|
"mean_token_accuracy": 0.15025533735752106,
|
|
"num_tokens": 3881276.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.357042551040649,
|
|
"epoch": 1.8388483025354534,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048737218052786275,
|
|
"loss": 6.2863,
|
|
"mean_token_accuracy": 0.14599718973040582,
|
|
"num_tokens": 3891610.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.469332885742188,
|
|
"epoch": 1.843145681134508,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048726212998104554,
|
|
"loss": 6.2036,
|
|
"mean_token_accuracy": 0.1476315975189209,
|
|
"num_tokens": 3900584.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.3720924854278564,
|
|
"epoch": 1.8474430597335625,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004871516159385768,
|
|
"loss": 6.1288,
|
|
"mean_token_accuracy": 0.15351544842123985,
|
|
"num_tokens": 3910208.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.2077491760253904,
|
|
"epoch": 1.8517404383326173,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004870406386417752,
|
|
"loss": 6.0609,
|
|
"mean_token_accuracy": 0.16224084943532943,
|
|
"num_tokens": 3918424.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.278759956359863,
|
|
"epoch": 1.8560378169316718,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004869291983329707,
|
|
"loss": 5.9946,
|
|
"mean_token_accuracy": 0.16720272302627565,
|
|
"num_tokens": 3926206.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.399888753890991,
|
|
"epoch": 1.8603351955307263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004868172952555044,
|
|
"loss": 6.0991,
|
|
"mean_token_accuracy": 0.1470145635306835,
|
|
"num_tokens": 3935769.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.315070724487304,
|
|
"epoch": 1.864632574129781,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004867049296537278,
|
|
"loss": 6.0903,
|
|
"mean_token_accuracy": 0.15458065569400786,
|
|
"num_tokens": 3945118.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.344206809997559,
|
|
"epoch": 1.8689299527288354,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004865921017730027,
|
|
"loss": 6.1791,
|
|
"mean_token_accuracy": 0.15464479327201844,
|
|
"num_tokens": 3954012.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.429828739166259,
|
|
"epoch": 1.87322733132789,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048647881185969995,
|
|
"loss": 6.1931,
|
|
"mean_token_accuracy": 0.14908381700515747,
|
|
"num_tokens": 3964239.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.33826584815979,
|
|
"epoch": 1.8775247099269445,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004863650601611994,
|
|
"loss": 6.0996,
|
|
"mean_token_accuracy": 0.1615213319659233,
|
|
"num_tokens": 3973694.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.327886295318604,
|
|
"epoch": 1.881822088525999,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048625084692588937,
|
|
"loss": 6.1415,
|
|
"mean_token_accuracy": 0.1605108693242073,
|
|
"num_tokens": 3982706.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.265936231613159,
|
|
"epoch": 1.8861194671250536,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048613617240316593,
|
|
"loss": 6.0825,
|
|
"mean_token_accuracy": 0.15816196352243422,
|
|
"num_tokens": 3990934.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.391205978393555,
|
|
"epoch": 1.8904168457241082,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004860210368434323,
|
|
"loss": 6.1513,
|
|
"mean_token_accuracy": 0.15758474171161652,
|
|
"num_tokens": 3999864.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.334363603591919,
|
|
"epoch": 1.8947142243231627,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048590544049809857,
|
|
"loss": 6.1514,
|
|
"mean_token_accuracy": 0.15803639888763427,
|
|
"num_tokens": 4008273.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.388755893707275,
|
|
"epoch": 1.8990116029222175,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000485789383619581,
|
|
"loss": 6.1719,
|
|
"mean_token_accuracy": 0.15583823770284652,
|
|
"num_tokens": 4017697.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.345981502532959,
|
|
"epoch": 1.903308981521272,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004856728664613015,
|
|
"loss": 6.1881,
|
|
"mean_token_accuracy": 0.14975374042987824,
|
|
"num_tokens": 4026775.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.3457728862762455,
|
|
"epoch": 1.9076063601203266,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048555588927768674,
|
|
"loss": 6.1523,
|
|
"mean_token_accuracy": 0.15700841099023818,
|
|
"num_tokens": 4036476.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.4198205947875975,
|
|
"epoch": 1.9119037387193811,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004854384523241683,
|
|
"loss": 6.1336,
|
|
"mean_token_accuracy": 0.1571663163602352,
|
|
"num_tokens": 4045221.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.218144416809082,
|
|
"epoch": 1.916201117318436,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048532055585718143,
|
|
"loss": 6.0619,
|
|
"mean_token_accuracy": 0.15748531818389894,
|
|
"num_tokens": 4053754.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.358814668655396,
|
|
"epoch": 1.9204984959174904,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048520220013416505,
|
|
"loss": 6.103,
|
|
"mean_token_accuracy": 0.1605447456240654,
|
|
"num_tokens": 4061730.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.3438163757324215,
|
|
"epoch": 1.924795874516545,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004850833854135607,
|
|
"loss": 6.1491,
|
|
"mean_token_accuracy": 0.15683530494570733,
|
|
"num_tokens": 4070501.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.367244625091553,
|
|
"epoch": 1.9290932531155995,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004849641119548122,
|
|
"loss": 6.2334,
|
|
"mean_token_accuracy": 0.14961420446634294,
|
|
"num_tokens": 4079621.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.398000574111938,
|
|
"epoch": 1.933390631714654,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000484844380018365,
|
|
"loss": 6.2167,
|
|
"mean_token_accuracy": 0.15164064317941667,
|
|
"num_tokens": 4090106.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.375770425796508,
|
|
"epoch": 1.9376880103137086,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000484724189865666,
|
|
"loss": 6.1578,
|
|
"mean_token_accuracy": 0.15246021896600723,
|
|
"num_tokens": 4099269.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.217796373367309,
|
|
"epoch": 1.9419853889127632,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004846035417591624,
|
|
"loss": 6.0917,
|
|
"mean_token_accuracy": 0.15897612571716307,
|
|
"num_tokens": 4108414.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.376744508743286,
|
|
"epoch": 1.9462827675118177,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004844824359623014,
|
|
"loss": 6.2234,
|
|
"mean_token_accuracy": 0.14845603406429292,
|
|
"num_tokens": 4117731.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.413218784332275,
|
|
"epoch": 1.9505801461108723,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048436087273952966,
|
|
"loss": 6.2001,
|
|
"mean_token_accuracy": 0.14851808845996856,
|
|
"num_tokens": 4127194.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.277564620971679,
|
|
"epoch": 1.9548775247099268,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048423885235629265,
|
|
"loss": 6.1488,
|
|
"mean_token_accuracy": 0.15824481397867202,
|
|
"num_tokens": 4135594.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.359572219848633,
|
|
"epoch": 1.9591749033089814,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004841163750790342,
|
|
"loss": 6.1804,
|
|
"mean_token_accuracy": 0.15617654621601104,
|
|
"num_tokens": 4145027.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.301140403747558,
|
|
"epoch": 1.9634722819080361,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048399344117519555,
|
|
"loss": 6.0431,
|
|
"mean_token_accuracy": 0.15682056695222854,
|
|
"num_tokens": 4153754.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.266417598724365,
|
|
"epoch": 1.9677696605070907,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048387005091321544,
|
|
"loss": 6.1066,
|
|
"mean_token_accuracy": 0.16042741984128953,
|
|
"num_tokens": 4162765.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.3823741436004635,
|
|
"epoch": 1.9720670391061452,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048374620456252877,
|
|
"loss": 6.1293,
|
|
"mean_token_accuracy": 0.15764901116490365,
|
|
"num_tokens": 4171589.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.2937760829925535,
|
|
"epoch": 1.9763644177052,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048362190239356644,
|
|
"loss": 6.1393,
|
|
"mean_token_accuracy": 0.15565742254257203,
|
|
"num_tokens": 4181817.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.305324840545654,
|
|
"epoch": 1.9806617963042545,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048349714467775474,
|
|
"loss": 6.0995,
|
|
"mean_token_accuracy": 0.14838732779026031,
|
|
"num_tokens": 4191350.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.278328514099121,
|
|
"epoch": 1.984959174903309,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048337193168751464,
|
|
"loss": 6.1486,
|
|
"mean_token_accuracy": 0.15034544318914414,
|
|
"num_tokens": 4199888.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.390921354293823,
|
|
"epoch": 1.9892565535023636,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004832462636962613,
|
|
"loss": 6.1298,
|
|
"mean_token_accuracy": 0.1492708593606949,
|
|
"num_tokens": 4209509.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.316125965118408,
|
|
"epoch": 1.9935539321014182,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004831201409784034,
|
|
"loss": 6.072,
|
|
"mean_token_accuracy": 0.1605883792042732,
|
|
"num_tokens": 4218496.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.255798053741455,
|
|
"epoch": 1.9978513107004727,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004829935638093424,
|
|
"loss": 6.1087,
|
|
"mean_token_accuracy": 0.15721286535263063,
|
|
"num_tokens": 4227504.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.355179150899251,
|
|
"epoch": 2.0017189514396216,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004828665324654724,
|
|
"loss": 6.0277,
|
|
"mean_token_accuracy": 0.15705766446060604,
|
|
"num_tokens": 4235338.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.347105932235718,
|
|
"epoch": 2.006016330038676,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004827390472241791,
|
|
"loss": 5.7915,
|
|
"mean_token_accuracy": 0.16444853693246841,
|
|
"num_tokens": 4244905.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.278535604476929,
|
|
"epoch": 2.010313708637731,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004826111083638392,
|
|
"loss": 5.8696,
|
|
"mean_token_accuracy": 0.16559881940484047,
|
|
"num_tokens": 4254533.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.316350841522217,
|
|
"epoch": 2.0146110872367857,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048248271616382,
|
|
"loss": 5.8463,
|
|
"mean_token_accuracy": 0.16606585830450057,
|
|
"num_tokens": 4264023.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.242398643493653,
|
|
"epoch": 2.0189084658358403,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048235387090447894,
|
|
"loss": 5.876,
|
|
"mean_token_accuracy": 0.16416406631469727,
|
|
"num_tokens": 4273298.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.315939903259277,
|
|
"epoch": 2.023205844434895,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048222457286716235,
|
|
"loss": 5.8197,
|
|
"mean_token_accuracy": 0.16837385147809983,
|
|
"num_tokens": 4283244.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.246707153320313,
|
|
"epoch": 2.0275032230339494,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048209482233420564,
|
|
"loss": 5.7698,
|
|
"mean_token_accuracy": 0.17663903087377547,
|
|
"num_tokens": 4291677.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.277777862548828,
|
|
"epoch": 2.031800601633004,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000481964619588932,
|
|
"loss": 5.8196,
|
|
"mean_token_accuracy": 0.170748533308506,
|
|
"num_tokens": 4300822.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.263119220733643,
|
|
"epoch": 2.0360979802320585,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004818339649156523,
|
|
"loss": 5.8415,
|
|
"mean_token_accuracy": 0.17079205960035324,
|
|
"num_tokens": 4310149.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.12338604927063,
|
|
"epoch": 2.040395358831113,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048170285859966395,
|
|
"loss": 5.7423,
|
|
"mean_token_accuracy": 0.17794516831636428,
|
|
"num_tokens": 4319109.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.281350469589233,
|
|
"epoch": 2.0446927374301676,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048157130092725087,
|
|
"loss": 5.7302,
|
|
"mean_token_accuracy": 0.17352935224771499,
|
|
"num_tokens": 4327921.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.251600027084351,
|
|
"epoch": 2.048990116029222,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004814392921856824,
|
|
"loss": 5.8821,
|
|
"mean_token_accuracy": 0.17077834606170655,
|
|
"num_tokens": 4338026.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.208239459991455,
|
|
"epoch": 2.0532874946282766,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004813068326632128,
|
|
"loss": 5.7272,
|
|
"mean_token_accuracy": 0.17698098421096803,
|
|
"num_tokens": 4347794.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.288045644760132,
|
|
"epoch": 2.057584873227331,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004811739226490809,
|
|
"loss": 5.917,
|
|
"mean_token_accuracy": 0.16798364371061325,
|
|
"num_tokens": 4357249.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.22069525718689,
|
|
"epoch": 2.0618822518263857,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048104056243350896,
|
|
"loss": 5.8434,
|
|
"mean_token_accuracy": 0.16641683727502823,
|
|
"num_tokens": 4366053.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.237291955947876,
|
|
"epoch": 2.0661796304254403,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004809067523077023,
|
|
"loss": 5.8614,
|
|
"mean_token_accuracy": 0.1700182244181633,
|
|
"num_tokens": 4375543.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.230785751342774,
|
|
"epoch": 2.0704770090244953,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048077249256384884,
|
|
"loss": 5.7564,
|
|
"mean_token_accuracy": 0.17438603639602662,
|
|
"num_tokens": 4384332.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.19397988319397,
|
|
"epoch": 2.07477438762355,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004806377834951182,
|
|
"loss": 5.8466,
|
|
"mean_token_accuracy": 0.1659790098667145,
|
|
"num_tokens": 4393670.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.280642461776734,
|
|
"epoch": 2.0790717662226044,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048050262539566104,
|
|
"loss": 5.8543,
|
|
"mean_token_accuracy": 0.17281297594308853,
|
|
"num_tokens": 4402763.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.234371662139893,
|
|
"epoch": 2.083369144821659,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004803670185606087,
|
|
"loss": 5.766,
|
|
"mean_token_accuracy": 0.17708782404661177,
|
|
"num_tokens": 4411863.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.2064672946929935,
|
|
"epoch": 2.0876665234207135,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004802309632860724,
|
|
"loss": 5.8476,
|
|
"mean_token_accuracy": 0.17201682031154633,
|
|
"num_tokens": 4421110.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.299275922775268,
|
|
"epoch": 2.091963902019768,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048009445986914236,
|
|
"loss": 5.8416,
|
|
"mean_token_accuracy": 0.165672005712986,
|
|
"num_tokens": 4430249.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.197122812271118,
|
|
"epoch": 2.0962612806188226,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047995750860788756,
|
|
"loss": 5.8269,
|
|
"mean_token_accuracy": 0.16112401485443115,
|
|
"num_tokens": 4439686.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.239287519454956,
|
|
"epoch": 2.100558659217877,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004798201098013547,
|
|
"loss": 5.8073,
|
|
"mean_token_accuracy": 0.17286145985126494,
|
|
"num_tokens": 4448645.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.17987093925476,
|
|
"epoch": 2.1048560378169316,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047968226374956797,
|
|
"loss": 5.785,
|
|
"mean_token_accuracy": 0.1676485523581505,
|
|
"num_tokens": 4456870.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.148839092254638,
|
|
"epoch": 2.109153416415986,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047954397075352794,
|
|
"loss": 5.804,
|
|
"mean_token_accuracy": 0.1790194794535637,
|
|
"num_tokens": 4466287.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.161268472671509,
|
|
"epoch": 2.1134507950150407,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047940523111521136,
|
|
"loss": 5.7069,
|
|
"mean_token_accuracy": 0.17733812779188157,
|
|
"num_tokens": 4474461.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.201317834854126,
|
|
"epoch": 2.1177481736140953,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004792660451375701,
|
|
"loss": 5.7722,
|
|
"mean_token_accuracy": 0.17279447317123414,
|
|
"num_tokens": 4483002.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.191226196289063,
|
|
"epoch": 2.12204555221315,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047912641312453064,
|
|
"loss": 5.7874,
|
|
"mean_token_accuracy": 0.1739989399909973,
|
|
"num_tokens": 4492061.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.240914964675904,
|
|
"epoch": 2.1263429308122044,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047898633538099363,
|
|
"loss": 5.8403,
|
|
"mean_token_accuracy": 0.16375242471694945,
|
|
"num_tokens": 4501829.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.218794155120849,
|
|
"epoch": 2.130640309411259,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004788458122128327,
|
|
"loss": 5.8683,
|
|
"mean_token_accuracy": 0.16395576894283295,
|
|
"num_tokens": 4511539.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.197416687011719,
|
|
"epoch": 2.134937688010314,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047870484392689434,
|
|
"loss": 5.7256,
|
|
"mean_token_accuracy": 0.1732256680727005,
|
|
"num_tokens": 4520425.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.153204393386841,
|
|
"epoch": 2.1392350666093685,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000478563430830997,
|
|
"loss": 5.8237,
|
|
"mean_token_accuracy": 0.1660924568772316,
|
|
"num_tokens": 4529474.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.238351488113404,
|
|
"epoch": 2.143532445208423,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047842157323393035,
|
|
"loss": 5.7621,
|
|
"mean_token_accuracy": 0.17213856130838395,
|
|
"num_tokens": 4538082.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.185346221923828,
|
|
"epoch": 2.1478298238074776,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004782792714454547,
|
|
"loss": 5.9584,
|
|
"mean_token_accuracy": 0.1624767854809761,
|
|
"num_tokens": 4547340.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.1478298238074776,
|
|
"eval_entropy": 5.975759233440365,
|
|
"eval_loss": 6.17199182510376,
|
|
"eval_mean_token_accuracy": 0.15897784858673542,
|
|
"eval_num_tokens": 4547340.0,
|
|
"eval_runtime": 2.0476,
|
|
"eval_samples_per_second": 1733.228,
|
|
"eval_steps_per_second": 216.837,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.180462265014649,
|
|
"epoch": 2.152127202406532,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004781365257763002,
|
|
"loss": 5.787,
|
|
"mean_token_accuracy": 0.17266686409711837,
|
|
"num_tokens": 4556415.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.12634973526001,
|
|
"epoch": 2.1564245810055866,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00047799333653816633,
|
|
"loss": 5.6887,
|
|
"mean_token_accuracy": 0.18215310871601104,
|
|
"num_tokens": 4565156.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.176189327239991,
|
|
"epoch": 2.160721959604641,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047784970404372124,
|
|
"loss": 5.7844,
|
|
"mean_token_accuracy": 0.1708789974451065,
|
|
"num_tokens": 4574678.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.102172183990478,
|
|
"epoch": 2.1650193382036957,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047770562860660083,
|
|
"loss": 5.811,
|
|
"mean_token_accuracy": 0.16656892001628876,
|
|
"num_tokens": 4583253.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.217717409133911,
|
|
"epoch": 2.1693167168027503,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004775611105414083,
|
|
"loss": 5.88,
|
|
"mean_token_accuracy": 0.16319236159324646,
|
|
"num_tokens": 4594042.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.179096984863281,
|
|
"epoch": 2.173614095401805,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004774161501637133,
|
|
"loss": 5.8208,
|
|
"mean_token_accuracy": 0.16849268227815628,
|
|
"num_tokens": 4603128.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.120770788192749,
|
|
"epoch": 2.1779114740008594,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004772707477900514,
|
|
"loss": 5.805,
|
|
"mean_token_accuracy": 0.17015553265810013,
|
|
"num_tokens": 4611537.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.254914093017578,
|
|
"epoch": 2.182208852599914,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004771249037379232,
|
|
"loss": 5.8984,
|
|
"mean_token_accuracy": 0.16680168211460114,
|
|
"num_tokens": 4622481.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.1392961025238035,
|
|
"epoch": 2.1865062311989685,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004769786183257939,
|
|
"loss": 5.7985,
|
|
"mean_token_accuracy": 0.17476972788572312,
|
|
"num_tokens": 4631259.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.153097438812256,
|
|
"epoch": 2.190803609798023,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004768318918730924,
|
|
"loss": 5.7586,
|
|
"mean_token_accuracy": 0.17578112632036208,
|
|
"num_tokens": 4640266.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.1481832504272464,
|
|
"epoch": 2.195100988397078,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047668472470021044,
|
|
"loss": 5.8113,
|
|
"mean_token_accuracy": 0.16784311085939407,
|
|
"num_tokens": 4649520.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.236191320419311,
|
|
"epoch": 2.1993983669961326,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004765371171285025,
|
|
"loss": 5.7573,
|
|
"mean_token_accuracy": 0.17908186167478563,
|
|
"num_tokens": 4658501.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.072221088409424,
|
|
"epoch": 2.203695745595187,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047638906948028445,
|
|
"loss": 5.8192,
|
|
"mean_token_accuracy": 0.16885218769311905,
|
|
"num_tokens": 4667567.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.158999061584472,
|
|
"epoch": 2.2079931241942417,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047624058207883317,
|
|
"loss": 5.8154,
|
|
"mean_token_accuracy": 0.16924346387386321,
|
|
"num_tokens": 4676618.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.275905895233154,
|
|
"epoch": 2.212290502793296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047609165524838576,
|
|
"loss": 5.8748,
|
|
"mean_token_accuracy": 0.16927455067634584,
|
|
"num_tokens": 4685967.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.100191926956176,
|
|
"epoch": 2.2165878813923507,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004759422893141389,
|
|
"loss": 5.766,
|
|
"mean_token_accuracy": 0.17206043750047684,
|
|
"num_tokens": 4694568.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.1515192031860355,
|
|
"epoch": 2.2208852599914053,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004757924846022482,
|
|
"loss": 5.826,
|
|
"mean_token_accuracy": 0.17090158760547638,
|
|
"num_tokens": 4703648.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.158021259307861,
|
|
"epoch": 2.22518263859046,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047564224143982714,
|
|
"loss": 5.6863,
|
|
"mean_token_accuracy": 0.1804699569940567,
|
|
"num_tokens": 4712444.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.184865522384643,
|
|
"epoch": 2.2294800171895144,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047549156015494676,
|
|
"loss": 5.8404,
|
|
"mean_token_accuracy": 0.1678527906537056,
|
|
"num_tokens": 4722034.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.146687793731689,
|
|
"epoch": 2.233777395788569,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047534044107663484,
|
|
"loss": 5.8616,
|
|
"mean_token_accuracy": 0.1660257250070572,
|
|
"num_tokens": 4731344.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.193604421615601,
|
|
"epoch": 2.2380747743876235,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047518888453487496,
|
|
"loss": 5.7742,
|
|
"mean_token_accuracy": 0.18199435770511627,
|
|
"num_tokens": 4739302.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.141710472106934,
|
|
"epoch": 2.242372152986678,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004750368908606061,
|
|
"loss": 5.8788,
|
|
"mean_token_accuracy": 0.16633899062871932,
|
|
"num_tokens": 4748848.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.243501758575439,
|
|
"epoch": 2.2466695315857326,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047488446038572164,
|
|
"loss": 5.9385,
|
|
"mean_token_accuracy": 0.16254912167787552,
|
|
"num_tokens": 4758194.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.16309700012207,
|
|
"epoch": 2.250966910184787,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004747315934430688,
|
|
"loss": 5.8537,
|
|
"mean_token_accuracy": 0.16617514342069625,
|
|
"num_tokens": 4768081.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.1246990203857425,
|
|
"epoch": 2.2552642887838417,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000474578290366448,
|
|
"loss": 5.7741,
|
|
"mean_token_accuracy": 0.1719846934080124,
|
|
"num_tokens": 4776471.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.14447546005249,
|
|
"epoch": 2.259561667382896,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004744245514906117,
|
|
"loss": 5.7832,
|
|
"mean_token_accuracy": 0.17195819914340973,
|
|
"num_tokens": 4784403.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.066154432296753,
|
|
"epoch": 2.263859045981951,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047427037715126426,
|
|
"loss": 5.7561,
|
|
"mean_token_accuracy": 0.17391669005155563,
|
|
"num_tokens": 4792779.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.112756013870239,
|
|
"epoch": 2.2681564245810057,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004741157676850608,
|
|
"loss": 5.7404,
|
|
"mean_token_accuracy": 0.17708691954612732,
|
|
"num_tokens": 4801426.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.151607942581177,
|
|
"epoch": 2.2724538031800603,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047396072342960663,
|
|
"loss": 5.7871,
|
|
"mean_token_accuracy": 0.16531864404678345,
|
|
"num_tokens": 4810329.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.167152023315429,
|
|
"epoch": 2.276751181779115,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047380524472345645,
|
|
"loss": 5.8465,
|
|
"mean_token_accuracy": 0.1671397715806961,
|
|
"num_tokens": 4819544.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.143870735168457,
|
|
"epoch": 2.2810485603781694,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004736493319061134,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.16567971110343932,
|
|
"num_tokens": 4828113.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.107415771484375,
|
|
"epoch": 2.285345938977224,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004734929853180291,
|
|
"loss": 5.8272,
|
|
"mean_token_accuracy": 0.1672320321202278,
|
|
"num_tokens": 4836989.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.207330274581909,
|
|
"epoch": 2.2896433175762785,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047333620530060175,
|
|
"loss": 5.8623,
|
|
"mean_token_accuracy": 0.16721852421760558,
|
|
"num_tokens": 4847103.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.162304496765136,
|
|
"epoch": 2.293940696175333,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004731789921961764,
|
|
"loss": 5.8736,
|
|
"mean_token_accuracy": 0.17004917562007904,
|
|
"num_tokens": 4856238.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.193357849121094,
|
|
"epoch": 2.2982380747743876,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004730213463480434,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.17748985737562178,
|
|
"num_tokens": 4864608.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.143442440032959,
|
|
"epoch": 2.302535453373442,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047286326810043857,
|
|
"loss": 5.7374,
|
|
"mean_token_accuracy": 0.17498592138290406,
|
|
"num_tokens": 4873889.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.103658771514892,
|
|
"epoch": 2.3068328319724967,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047270475779854137,
|
|
"loss": 5.7804,
|
|
"mean_token_accuracy": 0.1758432075381279,
|
|
"num_tokens": 4882902.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.238151502609253,
|
|
"epoch": 2.311130210571551,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047254581578847507,
|
|
"loss": 5.7985,
|
|
"mean_token_accuracy": 0.16956950426101686,
|
|
"num_tokens": 4892390.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.069392299652099,
|
|
"epoch": 2.3154275891706058,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004723864424173055,
|
|
"loss": 5.9313,
|
|
"mean_token_accuracy": 0.16988434195518493,
|
|
"num_tokens": 4901625.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.1605274200439455,
|
|
"epoch": 2.3197249677696608,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004722266380330403,
|
|
"loss": 5.7187,
|
|
"mean_token_accuracy": 0.18315426409244537,
|
|
"num_tokens": 4910804.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.102558517456055,
|
|
"epoch": 2.3240223463687153,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047206640298462857,
|
|
"loss": 5.8022,
|
|
"mean_token_accuracy": 0.1743517056107521,
|
|
"num_tokens": 4920441.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.1133277893066404,
|
|
"epoch": 2.32831972496777,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047190573762195945,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.17087312042713165,
|
|
"num_tokens": 4930204.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.14404034614563,
|
|
"epoch": 2.3326171035668244,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047174464229586186,
|
|
"loss": 5.9433,
|
|
"mean_token_accuracy": 0.16308000683784485,
|
|
"num_tokens": 4941191.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.279735326766968,
|
|
"epoch": 2.336914482165879,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004715831173581036,
|
|
"loss": 5.9209,
|
|
"mean_token_accuracy": 0.16662475615739822,
|
|
"num_tokens": 4951825.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.089169502258301,
|
|
"epoch": 2.3412118607649335,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047142116316139073,
|
|
"loss": 5.8464,
|
|
"mean_token_accuracy": 0.1715902417898178,
|
|
"num_tokens": 4960632.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.170309162139892,
|
|
"epoch": 2.345509239363988,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004712587800593663,
|
|
"loss": 5.8798,
|
|
"mean_token_accuracy": 0.16993365585803985,
|
|
"num_tokens": 4969455.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.12351393699646,
|
|
"epoch": 2.3498066179630426,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004710959684066102,
|
|
"loss": 5.7859,
|
|
"mean_token_accuracy": 0.1763747364282608,
|
|
"num_tokens": 4978997.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.158613014221191,
|
|
"epoch": 2.354103996562097,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047093272855863803,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.17122582197189332,
|
|
"num_tokens": 4988305.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.098289728164673,
|
|
"epoch": 2.3584013751611517,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004707690608719003,
|
|
"loss": 5.7826,
|
|
"mean_token_accuracy": 0.1741472825407982,
|
|
"num_tokens": 4997022.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.148240280151367,
|
|
"epoch": 2.362698753760206,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004706049657037818,
|
|
"loss": 5.8468,
|
|
"mean_token_accuracy": 0.1676038146018982,
|
|
"num_tokens": 5005664.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.128693151473999,
|
|
"epoch": 2.3669961323592608,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004704404434126009,
|
|
"loss": 5.7993,
|
|
"mean_token_accuracy": 0.1630128264427185,
|
|
"num_tokens": 5014769.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.173029994964599,
|
|
"epoch": 2.3712935109583153,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047027549435760843,
|
|
"loss": 5.869,
|
|
"mean_token_accuracy": 0.16782066822052003,
|
|
"num_tokens": 5024060.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.2035542011260985,
|
|
"epoch": 2.37559088955737,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004701101188989872,
|
|
"loss": 5.9029,
|
|
"mean_token_accuracy": 0.16563379466533662,
|
|
"num_tokens": 5033046.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.131078624725342,
|
|
"epoch": 2.3798882681564244,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046994431739785114,
|
|
"loss": 5.7511,
|
|
"mean_token_accuracy": 0.18578273057937622,
|
|
"num_tokens": 5040894.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.131838798522949,
|
|
"epoch": 2.384185646755479,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046977809021624454,
|
|
"loss": 5.9175,
|
|
"mean_token_accuracy": 0.16830566823482512,
|
|
"num_tokens": 5050961.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.217153167724609,
|
|
"epoch": 2.3884830253545335,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004696114377171409,
|
|
"loss": 5.8447,
|
|
"mean_token_accuracy": 0.166962693631649,
|
|
"num_tokens": 5060226.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.121598243713379,
|
|
"epoch": 2.3927804039535885,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004694443602644429,
|
|
"loss": 5.8063,
|
|
"mean_token_accuracy": 0.16657978445291519,
|
|
"num_tokens": 5069225.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.101100969314575,
|
|
"epoch": 2.397077782552643,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004692768582229808,
|
|
"loss": 5.7858,
|
|
"mean_token_accuracy": 0.17338948845863342,
|
|
"num_tokens": 5078386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.113597059249878,
|
|
"epoch": 2.4013751611516976,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046910893195851213,
|
|
"loss": 5.7198,
|
|
"mean_token_accuracy": 0.1726340189576149,
|
|
"num_tokens": 5087161.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.0919578075408936,
|
|
"epoch": 2.405672539750752,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046894058183772074,
|
|
"loss": 5.892,
|
|
"mean_token_accuracy": 0.16800693273544312,
|
|
"num_tokens": 5096613.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.136939573287964,
|
|
"epoch": 2.4099699183498067,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000468771808228216,
|
|
"loss": 5.8324,
|
|
"mean_token_accuracy": 0.16829856783151625,
|
|
"num_tokens": 5106534.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.107866430282593,
|
|
"epoch": 2.414267296948861,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046860261149853197,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.16899872124195098,
|
|
"num_tokens": 5115975.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.077492666244507,
|
|
"epoch": 2.4185646755479158,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004684329920181268,
|
|
"loss": 5.751,
|
|
"mean_token_accuracy": 0.1743526503443718,
|
|
"num_tokens": 5124635.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.131529140472412,
|
|
"epoch": 2.4228620541469703,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046826295015738154,
|
|
"loss": 5.7277,
|
|
"mean_token_accuracy": 0.18355693519115449,
|
|
"num_tokens": 5133226.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.0389073371887205,
|
|
"epoch": 2.427159432746025,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004680924862875996,
|
|
"loss": 5.8249,
|
|
"mean_token_accuracy": 0.17242621779441833,
|
|
"num_tokens": 5142257.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.1347551345825195,
|
|
"epoch": 2.4314568113450794,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046792160078100605,
|
|
"loss": 5.803,
|
|
"mean_token_accuracy": 0.1744050070643425,
|
|
"num_tokens": 5150752.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.121770191192627,
|
|
"epoch": 2.435754189944134,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046775029401074653,
|
|
"loss": 5.7301,
|
|
"mean_token_accuracy": 0.18050158619880677,
|
|
"num_tokens": 5160237.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.1292938709259035,
|
|
"epoch": 2.4400515685431885,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046757856635088645,
|
|
"loss": 5.8006,
|
|
"mean_token_accuracy": 0.1770539328455925,
|
|
"num_tokens": 5169752.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.102383279800415,
|
|
"epoch": 2.444348947142243,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004674064181764105,
|
|
"loss": 5.8474,
|
|
"mean_token_accuracy": 0.17131757587194443,
|
|
"num_tokens": 5178892.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.149413537979126,
|
|
"epoch": 2.448646325741298,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046723384986322147,
|
|
"loss": 5.8235,
|
|
"mean_token_accuracy": 0.1708175078034401,
|
|
"num_tokens": 5188468.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.077389192581177,
|
|
"epoch": 2.4529437043403526,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004670608617881395,
|
|
"loss": 5.7566,
|
|
"mean_token_accuracy": 0.17592367827892302,
|
|
"num_tokens": 5197565.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.030880689620972,
|
|
"epoch": 2.457241082939407,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004668874543289014,
|
|
"loss": 5.7533,
|
|
"mean_token_accuracy": 0.17984721809625626,
|
|
"num_tokens": 5205791.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.102017164230347,
|
|
"epoch": 2.4615384615384617,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046671362786415986,
|
|
"loss": 5.7546,
|
|
"mean_token_accuracy": 0.18463018238544465,
|
|
"num_tokens": 5214773.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.001052331924439,
|
|
"epoch": 2.465835840137516,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046653938277348237,
|
|
"loss": 5.7784,
|
|
"mean_token_accuracy": 0.178888800740242,
|
|
"num_tokens": 5223734.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.224011754989624,
|
|
"epoch": 2.4701332187365708,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004663647194373505,
|
|
"loss": 5.8493,
|
|
"mean_token_accuracy": 0.1654559224843979,
|
|
"num_tokens": 5231742.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.084821510314941,
|
|
"epoch": 2.4744305973356253,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046618963823715913,
|
|
"loss": 5.8114,
|
|
"mean_token_accuracy": 0.1739817038178444,
|
|
"num_tokens": 5241673.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.109079217910766,
|
|
"epoch": 2.47872797593468,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046601413955521575,
|
|
"loss": 5.7746,
|
|
"mean_token_accuracy": 0.1694352611899376,
|
|
"num_tokens": 5250082.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.092094850540161,
|
|
"epoch": 2.4830253545337344,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004658382237747393,
|
|
"loss": 5.8386,
|
|
"mean_token_accuracy": 0.1698906749486923,
|
|
"num_tokens": 5259680.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.122028970718384,
|
|
"epoch": 2.487322733132789,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046566189127985946,
|
|
"loss": 5.8246,
|
|
"mean_token_accuracy": 0.17455327808856963,
|
|
"num_tokens": 5269561.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.137590980529785,
|
|
"epoch": 2.4916201117318435,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000465485142455616,
|
|
"loss": 5.7746,
|
|
"mean_token_accuracy": 0.18081652075052262,
|
|
"num_tokens": 5278659.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.009606838226318,
|
|
"epoch": 2.495917490330898,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046530797768795765,
|
|
"loss": 5.7732,
|
|
"mean_token_accuracy": 0.180580236017704,
|
|
"num_tokens": 5287619.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.08758659362793,
|
|
"epoch": 2.5002148689299526,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046513039736374153,
|
|
"loss": 5.8748,
|
|
"mean_token_accuracy": 0.16648129373788834,
|
|
"num_tokens": 5297334.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.145037269592285,
|
|
"epoch": 2.504512247529007,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004649524018707319,
|
|
"loss": 5.8028,
|
|
"mean_token_accuracy": 0.17371988743543626,
|
|
"num_tokens": 5306208.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.061826229095459,
|
|
"epoch": 2.5088096261280617,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046477399159759996,
|
|
"loss": 5.7262,
|
|
"mean_token_accuracy": 0.17741942554712295,
|
|
"num_tokens": 5314754.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.948053169250488,
|
|
"epoch": 2.5131070047271162,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046459516693392246,
|
|
"loss": 5.751,
|
|
"mean_token_accuracy": 0.17220668345689774,
|
|
"num_tokens": 5324000.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.165147161483764,
|
|
"epoch": 2.517404383326171,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004644159282701808,
|
|
"loss": 5.7977,
|
|
"mean_token_accuracy": 0.17256153672933577,
|
|
"num_tokens": 5332478.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.164068746566772,
|
|
"epoch": 2.5217017619252258,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046423627599776076,
|
|
"loss": 5.881,
|
|
"mean_token_accuracy": 0.16446806192398072,
|
|
"num_tokens": 5341635.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.083116579055786,
|
|
"epoch": 2.5259991405242803,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000464056210508951,
|
|
"loss": 5.8562,
|
|
"mean_token_accuracy": 0.17011034190654756,
|
|
"num_tokens": 5350144.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.115471887588501,
|
|
"epoch": 2.530296519123335,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004638757321969426,
|
|
"loss": 5.7867,
|
|
"mean_token_accuracy": 0.17099311649799348,
|
|
"num_tokens": 5358788.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.1144256591796875,
|
|
"epoch": 2.5345938977223894,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046369484145582815,
|
|
"loss": 5.8724,
|
|
"mean_token_accuracy": 0.16514718383550644,
|
|
"num_tokens": 5368057.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.012566184997558,
|
|
"epoch": 2.538891276321444,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046351353868060054,
|
|
"loss": 5.7114,
|
|
"mean_token_accuracy": 0.18013515770435334,
|
|
"num_tokens": 5376739.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.148663520812988,
|
|
"epoch": 2.5431886549204985,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046333182426715273,
|
|
"loss": 5.8321,
|
|
"mean_token_accuracy": 0.17200638502836227,
|
|
"num_tokens": 5385967.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.130731725692749,
|
|
"epoch": 2.547486033519553,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046314969861227626,
|
|
"loss": 5.8503,
|
|
"mean_token_accuracy": 0.16146385669708252,
|
|
"num_tokens": 5395192.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.0897301197052,
|
|
"epoch": 2.5517834121186076,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004629671621136608,
|
|
"loss": 5.8185,
|
|
"mean_token_accuracy": 0.1707317277789116,
|
|
"num_tokens": 5404694.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.1051109790802,
|
|
"epoch": 2.556080790717662,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004627842151698931,
|
|
"loss": 5.8208,
|
|
"mean_token_accuracy": 0.1680385336279869,
|
|
"num_tokens": 5413102.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.055868244171142,
|
|
"epoch": 2.5603781693167167,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046260085818045625,
|
|
"loss": 5.8548,
|
|
"mean_token_accuracy": 0.17059293240308762,
|
|
"num_tokens": 5423339.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.159391784667969,
|
|
"epoch": 2.5646755479157712,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004624170915457284,
|
|
"loss": 5.8092,
|
|
"mean_token_accuracy": 0.1737206295132637,
|
|
"num_tokens": 5432377.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.093228054046631,
|
|
"epoch": 2.5689729265148262,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046223291566698264,
|
|
"loss": 5.7337,
|
|
"mean_token_accuracy": 0.17590138092637062,
|
|
"num_tokens": 5441038.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.048309326171875,
|
|
"epoch": 2.5732703051138808,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004620483309463855,
|
|
"loss": 5.752,
|
|
"mean_token_accuracy": 0.18060761988162993,
|
|
"num_tokens": 5449557.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.118342542648316,
|
|
"epoch": 2.5775676837129353,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004618633377869961,
|
|
"loss": 5.8908,
|
|
"mean_token_accuracy": 0.17027267962694168,
|
|
"num_tokens": 5458931.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.5775676837129353,
|
|
"eval_entropy": 5.922512502283664,
|
|
"eval_loss": 6.089641094207764,
|
|
"eval_mean_token_accuracy": 0.16509396373084537,
|
|
"eval_num_tokens": 5458931.0,
|
|
"eval_runtime": 2.0449,
|
|
"eval_samples_per_second": 1735.564,
|
|
"eval_steps_per_second": 217.129,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.080532598495483,
|
|
"epoch": 2.58186506231199,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004616779365927656,
|
|
"loss": 5.715,
|
|
"mean_token_accuracy": 0.18329965472221374,
|
|
"num_tokens": 5468539.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.97086615562439,
|
|
"epoch": 2.5861624409110444,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004614921277685361,
|
|
"loss": 5.6562,
|
|
"mean_token_accuracy": 0.18940457850694656,
|
|
"num_tokens": 5475710.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.014264154434204,
|
|
"epoch": 2.590459819510099,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046130591172003976,
|
|
"loss": 5.8105,
|
|
"mean_token_accuracy": 0.17161315381526948,
|
|
"num_tokens": 5484597.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.152267932891846,
|
|
"epoch": 2.5947571981091535,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004611192888538981,
|
|
"loss": 5.8783,
|
|
"mean_token_accuracy": 0.16452959179878235,
|
|
"num_tokens": 5493213.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.160617208480835,
|
|
"epoch": 2.599054576708208,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046093225957762084,
|
|
"loss": 5.8774,
|
|
"mean_token_accuracy": 0.1677599936723709,
|
|
"num_tokens": 5502556.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.121158790588379,
|
|
"epoch": 2.6033519553072626,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004607448242996051,
|
|
"loss": 5.7783,
|
|
"mean_token_accuracy": 0.17667657732963563,
|
|
"num_tokens": 5511779.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.09659161567688,
|
|
"epoch": 2.607649333906317,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004605569834291347,
|
|
"loss": 5.7762,
|
|
"mean_token_accuracy": 0.17719159871339799,
|
|
"num_tokens": 5520836.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.045120048522949,
|
|
"epoch": 2.6119467125053717,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046036873737637904,
|
|
"loss": 5.7919,
|
|
"mean_token_accuracy": 0.1743898034095764,
|
|
"num_tokens": 5529285.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.032482624053955,
|
|
"epoch": 2.6162440911044262,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004601800865523921,
|
|
"loss": 5.795,
|
|
"mean_token_accuracy": 0.171583154797554,
|
|
"num_tokens": 5538160.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.106417560577393,
|
|
"epoch": 2.620541469703481,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045999103136911204,
|
|
"loss": 5.801,
|
|
"mean_token_accuracy": 0.16901974827051164,
|
|
"num_tokens": 5547355.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.082255268096924,
|
|
"epoch": 2.6248388483025353,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045980157223935965,
|
|
"loss": 5.8134,
|
|
"mean_token_accuracy": 0.17005283683538436,
|
|
"num_tokens": 5557299.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.022762966156006,
|
|
"epoch": 2.62913622690159,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045961170957683806,
|
|
"loss": 5.7335,
|
|
"mean_token_accuracy": 0.17649647146463393,
|
|
"num_tokens": 5565469.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.093750953674316,
|
|
"epoch": 2.6334336055006444,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045942144379613147,
|
|
"loss": 5.8526,
|
|
"mean_token_accuracy": 0.17166510373353958,
|
|
"num_tokens": 5574740.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.13080358505249,
|
|
"epoch": 2.637730984099699,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045923077531270426,
|
|
"loss": 5.8407,
|
|
"mean_token_accuracy": 0.1691308617591858,
|
|
"num_tokens": 5583438.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.099384212493897,
|
|
"epoch": 2.6420283626987535,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004590397045429001,
|
|
"loss": 5.8094,
|
|
"mean_token_accuracy": 0.17814612835645677,
|
|
"num_tokens": 5592389.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.034897947311402,
|
|
"epoch": 2.646325741297808,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045884823190394134,
|
|
"loss": 5.7097,
|
|
"mean_token_accuracy": 0.18135049045085908,
|
|
"num_tokens": 5601598.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.026404762268067,
|
|
"epoch": 2.650623119896863,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004586563578139275,
|
|
"loss": 5.7991,
|
|
"mean_token_accuracy": 0.16875589936971663,
|
|
"num_tokens": 5610498.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.030868768692017,
|
|
"epoch": 2.6549204984959176,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045846408269183505,
|
|
"loss": 5.7054,
|
|
"mean_token_accuracy": 0.1822717770934105,
|
|
"num_tokens": 5620082.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.124869680404663,
|
|
"epoch": 2.659217877094972,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045827140695751603,
|
|
"loss": 5.7925,
|
|
"mean_token_accuracy": 0.17561250925064087,
|
|
"num_tokens": 5630291.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.034296226501465,
|
|
"epoch": 2.6635152556940267,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004580783310316971,
|
|
"loss": 5.7735,
|
|
"mean_token_accuracy": 0.1745203271508217,
|
|
"num_tokens": 5638784.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.955547714233399,
|
|
"epoch": 2.6678126342930812,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045788485533597895,
|
|
"loss": 5.6462,
|
|
"mean_token_accuracy": 0.1883938804268837,
|
|
"num_tokens": 5647968.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.074917793273926,
|
|
"epoch": 2.672110012892136,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045769098029283526,
|
|
"loss": 5.8675,
|
|
"mean_token_accuracy": 0.16595781594514847,
|
|
"num_tokens": 5657543.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.098982810974121,
|
|
"epoch": 2.6764073914911903,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004574967063256115,
|
|
"loss": 5.7801,
|
|
"mean_token_accuracy": 0.17747018337249756,
|
|
"num_tokens": 5666535.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.071979904174805,
|
|
"epoch": 2.680704770090245,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045730203385852447,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.17044119387865067,
|
|
"num_tokens": 5676273.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.99071159362793,
|
|
"epoch": 2.6850021486892994,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000457106963316661,
|
|
"loss": 5.7612,
|
|
"mean_token_accuracy": 0.177787147462368,
|
|
"num_tokens": 5684888.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.075513315200806,
|
|
"epoch": 2.689299527288354,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045691149512597717,
|
|
"loss": 5.8228,
|
|
"mean_token_accuracy": 0.17054860144853592,
|
|
"num_tokens": 5693626.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.097319412231445,
|
|
"epoch": 2.6935969058874085,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045671562971329736,
|
|
"loss": 5.7252,
|
|
"mean_token_accuracy": 0.18006587252020836,
|
|
"num_tokens": 5702542.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.990765905380249,
|
|
"epoch": 2.6978942844864635,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045651936750631337,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.17453034669160844,
|
|
"num_tokens": 5711440.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.161540746688843,
|
|
"epoch": 2.702191663085518,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045632270893358333,
|
|
"loss": 5.8408,
|
|
"mean_token_accuracy": 0.17016429752111434,
|
|
"num_tokens": 5721495.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.113433980941773,
|
|
"epoch": 2.7064890416845726,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004561256544245312,
|
|
"loss": 5.8702,
|
|
"mean_token_accuracy": 0.16516700088977815,
|
|
"num_tokens": 5730664.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.01002688407898,
|
|
"epoch": 2.710786420283627,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000455928204409445,
|
|
"loss": 5.7398,
|
|
"mean_token_accuracy": 0.18062053769826888,
|
|
"num_tokens": 5740229.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.008193063735962,
|
|
"epoch": 2.7150837988826817,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045573035931947684,
|
|
"loss": 5.7378,
|
|
"mean_token_accuracy": 0.1744979053735733,
|
|
"num_tokens": 5748549.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.039789533615112,
|
|
"epoch": 2.7193811774817362,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004555321195866411,
|
|
"loss": 5.6806,
|
|
"mean_token_accuracy": 0.17732828259468078,
|
|
"num_tokens": 5757603.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.080458354949951,
|
|
"epoch": 2.723678556080791,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004553334856438143,
|
|
"loss": 5.8618,
|
|
"mean_token_accuracy": 0.16976358145475387,
|
|
"num_tokens": 5767520.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.09289813041687,
|
|
"epoch": 2.7279759346798453,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045513445792473356,
|
|
"loss": 5.8583,
|
|
"mean_token_accuracy": 0.1657076820731163,
|
|
"num_tokens": 5776778.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.129758882522583,
|
|
"epoch": 2.7322733132789,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004549350368639958,
|
|
"loss": 5.8808,
|
|
"mean_token_accuracy": 0.16651461273431778,
|
|
"num_tokens": 5785652.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.144531726837158,
|
|
"epoch": 2.7365706918779544,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045473522289705693,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.1734338730573654,
|
|
"num_tokens": 5795766.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.992430114746094,
|
|
"epoch": 2.740868070477009,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045453501646023085,
|
|
"loss": 5.8822,
|
|
"mean_token_accuracy": 0.16669443398714065,
|
|
"num_tokens": 5804504.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.04656753540039,
|
|
"epoch": 2.7451654490760635,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045433441799068837,
|
|
"loss": 5.7879,
|
|
"mean_token_accuracy": 0.17372047603130342,
|
|
"num_tokens": 5814161.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.094403076171875,
|
|
"epoch": 2.749462827675118,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004541334279264562,
|
|
"loss": 5.6942,
|
|
"mean_token_accuracy": 0.18637760877609252,
|
|
"num_tokens": 5822235.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.033037233352661,
|
|
"epoch": 2.7537602062741726,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045393204670641656,
|
|
"loss": 5.7009,
|
|
"mean_token_accuracy": 0.17470744848251343,
|
|
"num_tokens": 5831572.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.930374097824097,
|
|
"epoch": 2.758057584873227,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004537302747703055,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.18409457355737685,
|
|
"num_tokens": 5839694.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.124579620361328,
|
|
"epoch": 2.7623549634722817,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045352811255871216,
|
|
"loss": 5.8448,
|
|
"mean_token_accuracy": 0.17230847403407096,
|
|
"num_tokens": 5849131.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.174058246612549,
|
|
"epoch": 2.7666523420713363,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00045332556051307804,
|
|
"loss": 5.7711,
|
|
"mean_token_accuracy": 0.1720232903957367,
|
|
"num_tokens": 5858861.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.060689687728882,
|
|
"epoch": 2.770949720670391,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045312261907569585,
|
|
"loss": 5.7833,
|
|
"mean_token_accuracy": 0.17473076432943344,
|
|
"num_tokens": 5867585.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.015534782409668,
|
|
"epoch": 2.775247099269446,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045291928868970867,
|
|
"loss": 5.7865,
|
|
"mean_token_accuracy": 0.16985856741666794,
|
|
"num_tokens": 5876256.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.038353967666626,
|
|
"epoch": 2.7795444778685003,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004527155697991087,
|
|
"loss": 5.8471,
|
|
"mean_token_accuracy": 0.16595111042261124,
|
|
"num_tokens": 5885302.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.021387720108033,
|
|
"epoch": 2.783841856467555,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004525114628487365,
|
|
"loss": 5.8628,
|
|
"mean_token_accuracy": 0.16938397139310837,
|
|
"num_tokens": 5895066.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.151897144317627,
|
|
"epoch": 2.7881392350666094,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045230696828428026,
|
|
"loss": 5.8557,
|
|
"mean_token_accuracy": 0.16848236918449402,
|
|
"num_tokens": 5903258.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.039195203781128,
|
|
"epoch": 2.792436613665664,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004521020865522742,
|
|
"loss": 5.7511,
|
|
"mean_token_accuracy": 0.1710444927215576,
|
|
"num_tokens": 5911714.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.995623922348022,
|
|
"epoch": 2.7967339922647185,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045189681810009827,
|
|
"loss": 5.8176,
|
|
"mean_token_accuracy": 0.17098681181669234,
|
|
"num_tokens": 5920432.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.183896541595459,
|
|
"epoch": 2.801031370863773,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045169116337597653,
|
|
"loss": 5.8195,
|
|
"mean_token_accuracy": 0.1705167680978775,
|
|
"num_tokens": 5929202.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.115947484970093,
|
|
"epoch": 2.8053287494628276,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000451485122828977,
|
|
"loss": 5.8601,
|
|
"mean_token_accuracy": 0.16596955806016922,
|
|
"num_tokens": 5938034.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.939763784408569,
|
|
"epoch": 2.809626128061882,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045127869690900956,
|
|
"loss": 5.7097,
|
|
"mean_token_accuracy": 0.18104571253061294,
|
|
"num_tokens": 5946944.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.011037588119507,
|
|
"epoch": 2.8139235066609367,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045107188606682613,
|
|
"loss": 5.8219,
|
|
"mean_token_accuracy": 0.17439836859703065,
|
|
"num_tokens": 5956475.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.145265722274781,
|
|
"epoch": 2.8182208852599913,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004508646907540188,
|
|
"loss": 5.7788,
|
|
"mean_token_accuracy": 0.1669231042265892,
|
|
"num_tokens": 5965814.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.047088956832885,
|
|
"epoch": 2.8225182638590463,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004506571114230195,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.16392946541309356,
|
|
"num_tokens": 5973850.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.966979217529297,
|
|
"epoch": 2.826815642458101,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045044914852709824,
|
|
"loss": 5.7718,
|
|
"mean_token_accuracy": 0.17040073573589326,
|
|
"num_tokens": 5982987.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.124370384216308,
|
|
"epoch": 2.8311130210571553,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004502408025203631,
|
|
"loss": 5.7567,
|
|
"mean_token_accuracy": 0.18116641640663148,
|
|
"num_tokens": 5992227.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.060417222976684,
|
|
"epoch": 2.83541039965621,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004500320738577584,
|
|
"loss": 5.7373,
|
|
"mean_token_accuracy": 0.18134199529886247,
|
|
"num_tokens": 6000243.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.009725856781006,
|
|
"epoch": 2.8397077782552644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044982296299506407,
|
|
"loss": 5.7396,
|
|
"mean_token_accuracy": 0.1772996261715889,
|
|
"num_tokens": 6009771.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.065958309173584,
|
|
"epoch": 2.844005156854319,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004496134703888948,
|
|
"loss": 5.8227,
|
|
"mean_token_accuracy": 0.17212264090776444,
|
|
"num_tokens": 6018683.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.064363861083985,
|
|
"epoch": 2.8483025354533735,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044940359649669846,
|
|
"loss": 5.6744,
|
|
"mean_token_accuracy": 0.18227704763412475,
|
|
"num_tokens": 6027422.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.975715494155883,
|
|
"epoch": 2.852599914052428,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044919334177675595,
|
|
"loss": 5.7633,
|
|
"mean_token_accuracy": 0.1744269087910652,
|
|
"num_tokens": 6035670.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.036557149887085,
|
|
"epoch": 2.8568972926514826,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044898270668817955,
|
|
"loss": 5.6979,
|
|
"mean_token_accuracy": 0.1815047174692154,
|
|
"num_tokens": 6044092.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.020293951034546,
|
|
"epoch": 2.861194671250537,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000448771691690912,
|
|
"loss": 5.7773,
|
|
"mean_token_accuracy": 0.16777832806110382,
|
|
"num_tokens": 6053970.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.0354420185089115,
|
|
"epoch": 2.8654920498495917,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004485602972457257,
|
|
"loss": 5.7383,
|
|
"mean_token_accuracy": 0.1797216445207596,
|
|
"num_tokens": 6062965.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.068978786468506,
|
|
"epoch": 2.8697894284486463,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044834852381422165,
|
|
"loss": 5.8049,
|
|
"mean_token_accuracy": 0.17490418255329132,
|
|
"num_tokens": 6072420.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.005989837646484,
|
|
"epoch": 2.874086807047701,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044813637185882836,
|
|
"loss": 5.7175,
|
|
"mean_token_accuracy": 0.17540892213582993,
|
|
"num_tokens": 6080915.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.08394684791565,
|
|
"epoch": 2.8783841856467554,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044792384184280106,
|
|
"loss": 5.8546,
|
|
"mean_token_accuracy": 0.16469819992780685,
|
|
"num_tokens": 6090453.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.007422256469726,
|
|
"epoch": 2.88268156424581,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044771093423022013,
|
|
"loss": 5.8795,
|
|
"mean_token_accuracy": 0.16327449679374695,
|
|
"num_tokens": 6099390.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.039740133285522,
|
|
"epoch": 2.8869789428448644,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004474976494859909,
|
|
"loss": 5.798,
|
|
"mean_token_accuracy": 0.17494250684976578,
|
|
"num_tokens": 6108677.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.058880233764649,
|
|
"epoch": 2.891276321443919,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004472839880758419,
|
|
"loss": 5.716,
|
|
"mean_token_accuracy": 0.17659443318843843,
|
|
"num_tokens": 6117151.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.116930532455444,
|
|
"epoch": 2.8955737000429735,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004470699504663242,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.16731317192316056,
|
|
"num_tokens": 6127167.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.032348012924194,
|
|
"epoch": 2.899871078642028,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004468555371248104,
|
|
"loss": 5.7315,
|
|
"mean_token_accuracy": 0.1812448814511299,
|
|
"num_tokens": 6136487.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.042793130874633,
|
|
"epoch": 2.904168457241083,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004466407485194937,
|
|
"loss": 5.8432,
|
|
"mean_token_accuracy": 0.16948612183332443,
|
|
"num_tokens": 6145334.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.0380902767181395,
|
|
"epoch": 2.9084658358401376,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004464255851193864,
|
|
"loss": 5.7558,
|
|
"mean_token_accuracy": 0.17524855434894562,
|
|
"num_tokens": 6155062.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.0504677295684814,
|
|
"epoch": 2.912763214439192,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.0004462100473943194,
|
|
"loss": 5.6948,
|
|
"mean_token_accuracy": 0.1831045612692833,
|
|
"num_tokens": 6164313.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.009590721130371,
|
|
"epoch": 2.9170605930382467,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000445994135814941,
|
|
"loss": 5.7596,
|
|
"mean_token_accuracy": 0.17282265722751616,
|
|
"num_tokens": 6173513.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.019471836090088,
|
|
"epoch": 2.9213579716373013,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044577785085271566,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.17321082055568696,
|
|
"num_tokens": 6182000.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.081956481933593,
|
|
"epoch": 2.925655350236356,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004455611929799235,
|
|
"loss": 5.8084,
|
|
"mean_token_accuracy": 0.16455612033605577,
|
|
"num_tokens": 6191887.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.956260538101196,
|
|
"epoch": 2.9299527288354104,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004453441626696585,
|
|
"loss": 5.8554,
|
|
"mean_token_accuracy": 0.16420858800411225,
|
|
"num_tokens": 6202897.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.078850793838501,
|
|
"epoch": 2.934250107434465,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044512676039582823,
|
|
"loss": 5.7438,
|
|
"mean_token_accuracy": 0.18065994828939438,
|
|
"num_tokens": 6211811.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.1270510196685795,
|
|
"epoch": 2.9385474860335195,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004449089866331524,
|
|
"loss": 5.7475,
|
|
"mean_token_accuracy": 0.1779635578393936,
|
|
"num_tokens": 6219896.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.887205791473389,
|
|
"epoch": 2.942844864632574,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004446908418571617,
|
|
"loss": 5.737,
|
|
"mean_token_accuracy": 0.17775188386440277,
|
|
"num_tokens": 6228212.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.039950180053711,
|
|
"epoch": 2.9471422432316285,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004444723265441973,
|
|
"loss": 5.896,
|
|
"mean_token_accuracy": 0.16747722327709197,
|
|
"num_tokens": 6238133.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.076528787612915,
|
|
"epoch": 2.9514396218306835,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004442534411714092,
|
|
"loss": 5.7945,
|
|
"mean_token_accuracy": 0.16944347620010375,
|
|
"num_tokens": 6247331.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.101355123519897,
|
|
"epoch": 2.955737000429738,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044403418621675555,
|
|
"loss": 5.7926,
|
|
"mean_token_accuracy": 0.17052113264799118,
|
|
"num_tokens": 6255280.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.050349187850952,
|
|
"epoch": 2.9600343790287926,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004438145621590017,
|
|
"loss": 5.7555,
|
|
"mean_token_accuracy": 0.17491218596696853,
|
|
"num_tokens": 6264752.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.9742930889129635,
|
|
"epoch": 2.964331757627847,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044359456947771857,
|
|
"loss": 5.7023,
|
|
"mean_token_accuracy": 0.17401786297559738,
|
|
"num_tokens": 6273258.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.869597768783569,
|
|
"epoch": 2.9686291362269017,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004433742086532824,
|
|
"loss": 5.6228,
|
|
"mean_token_accuracy": 0.19265587478876114,
|
|
"num_tokens": 6281584.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.010894346237182,
|
|
"epoch": 2.9729265148259563,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044315348016687317,
|
|
"loss": 5.7472,
|
|
"mean_token_accuracy": 0.17217940390110015,
|
|
"num_tokens": 6290016.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.982230138778687,
|
|
"epoch": 2.977223893425011,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004429323845004736,
|
|
"loss": 5.6523,
|
|
"mean_token_accuracy": 0.18324829190969466,
|
|
"num_tokens": 6298569.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.025563192367554,
|
|
"epoch": 2.9815212720240654,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044271092213686824,
|
|
"loss": 5.6855,
|
|
"mean_token_accuracy": 0.18166320472955705,
|
|
"num_tokens": 6307684.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.1143230438232425,
|
|
"epoch": 2.98581865062312,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044248909355964247,
|
|
"loss": 5.8192,
|
|
"mean_token_accuracy": 0.17195742726325988,
|
|
"num_tokens": 6317767.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.101650047302246,
|
|
"epoch": 2.9901160292221745,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044226689925318117,
|
|
"loss": 5.8454,
|
|
"mean_token_accuracy": 0.16614989936351776,
|
|
"num_tokens": 6327457.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.9672809600830075,
|
|
"epoch": 2.994413407821229,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044204433970266785,
|
|
"loss": 5.6491,
|
|
"mean_token_accuracy": 0.1888262301683426,
|
|
"num_tokens": 6335747.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.983137941360473,
|
|
"epoch": 2.9987107864202835,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004418214153940837,
|
|
"loss": 5.7429,
|
|
"mean_token_accuracy": 0.18020158410072326,
|
|
"num_tokens": 6344750.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.069730917612712,
|
|
"epoch": 3.002578427159433,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00044159812681420624,
|
|
"loss": 5.6774,
|
|
"mean_token_accuracy": 0.18010619613859388,
|
|
"num_tokens": 6354779.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.04861307144165,
|
|
"epoch": 3.0068758057584875,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004413744744506086,
|
|
"loss": 5.4671,
|
|
"mean_token_accuracy": 0.18785551339387893,
|
|
"num_tokens": 6363809.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.0068758057584875,
|
|
"eval_entropy": 5.758008357640859,
|
|
"eval_loss": 6.010996341705322,
|
|
"eval_mean_token_accuracy": 0.17076294478196818,
|
|
"eval_num_tokens": 6363809.0,
|
|
"eval_runtime": 2.0461,
|
|
"eval_samples_per_second": 1734.535,
|
|
"eval_steps_per_second": 217.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.996388673782349,
|
|
"epoch": 3.011173184357542,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044115045879165806,
|
|
"loss": 5.5232,
|
|
"mean_token_accuracy": 0.18338106274604798,
|
|
"num_tokens": 6373082.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.007624244689941,
|
|
"epoch": 3.0154705629565965,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044092608032651515,
|
|
"loss": 5.4884,
|
|
"mean_token_accuracy": 0.18493928611278534,
|
|
"num_tokens": 6381286.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.025672864913941,
|
|
"epoch": 3.019767941555651,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044070133954513305,
|
|
"loss": 5.4331,
|
|
"mean_token_accuracy": 0.19577560871839522,
|
|
"num_tokens": 6390217.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.000850200653076,
|
|
"epoch": 3.0240653201547056,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004404762369382555,
|
|
"loss": 5.4683,
|
|
"mean_token_accuracy": 0.18827376067638396,
|
|
"num_tokens": 6399276.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.93954758644104,
|
|
"epoch": 3.02836269875376,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044025077299741683,
|
|
"loss": 5.4445,
|
|
"mean_token_accuracy": 0.1970704421401024,
|
|
"num_tokens": 6407981.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.957830190658569,
|
|
"epoch": 3.0326600773528147,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044002494821494007,
|
|
"loss": 5.4438,
|
|
"mean_token_accuracy": 0.1922784000635147,
|
|
"num_tokens": 6416159.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.9005261898040775,
|
|
"epoch": 3.0369574559518693,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043979876308393635,
|
|
"loss": 5.4964,
|
|
"mean_token_accuracy": 0.19178588539361954,
|
|
"num_tokens": 6424564.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.050303220748901,
|
|
"epoch": 3.041254834550924,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004395722180983036,
|
|
"loss": 5.5298,
|
|
"mean_token_accuracy": 0.1850397542119026,
|
|
"num_tokens": 6434163.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.894874811172485,
|
|
"epoch": 3.0455522131499784,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043934531375272535,
|
|
"loss": 5.3505,
|
|
"mean_token_accuracy": 0.20217433124780654,
|
|
"num_tokens": 6443372.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.910711765289307,
|
|
"epoch": 3.049849591749033,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043911805054267015,
|
|
"loss": 5.4569,
|
|
"mean_token_accuracy": 0.19326651990413665,
|
|
"num_tokens": 6452638.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.0943183422088625,
|
|
"epoch": 3.0541469703480875,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043889042896439004,
|
|
"loss": 5.4504,
|
|
"mean_token_accuracy": 0.19366029798984527,
|
|
"num_tokens": 6461319.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.933924388885498,
|
|
"epoch": 3.0584443489471425,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00043866244951491946,
|
|
"loss": 5.3807,
|
|
"mean_token_accuracy": 0.20453428626060485,
|
|
"num_tokens": 6469506.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.953407573699951,
|
|
"epoch": 3.062741727546197,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043843411269207445,
|
|
"loss": 5.437,
|
|
"mean_token_accuracy": 0.1967134654521942,
|
|
"num_tokens": 6478404.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.93691873550415,
|
|
"epoch": 3.0670391061452515,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004382054189944514,
|
|
"loss": 5.3816,
|
|
"mean_token_accuracy": 0.19278321117162706,
|
|
"num_tokens": 6487447.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.878192138671875,
|
|
"epoch": 3.071336484744306,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004379763689214259,
|
|
"loss": 5.4196,
|
|
"mean_token_accuracy": 0.18803493976593016,
|
|
"num_tokens": 6496738.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.99528021812439,
|
|
"epoch": 3.0756338633433606,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004377469629731518,
|
|
"loss": 5.4317,
|
|
"mean_token_accuracy": 0.193548683822155,
|
|
"num_tokens": 6505848.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.963389873504639,
|
|
"epoch": 3.079931241942415,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004375172016505599,
|
|
"loss": 5.4138,
|
|
"mean_token_accuracy": 0.19534891694784165,
|
|
"num_tokens": 6515731.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.967396306991577,
|
|
"epoch": 3.0842286205414697,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004372870854553572,
|
|
"loss": 5.4706,
|
|
"mean_token_accuracy": 0.194082772731781,
|
|
"num_tokens": 6524914.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.949575996398925,
|
|
"epoch": 3.0885259991405243,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004370566148900255,
|
|
"loss": 5.4527,
|
|
"mean_token_accuracy": 0.19753452241420746,
|
|
"num_tokens": 6533712.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.982216501235962,
|
|
"epoch": 3.092823377739579,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043682579045782024,
|
|
"loss": 5.5375,
|
|
"mean_token_accuracy": 0.18995364159345626,
|
|
"num_tokens": 6543313.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.923231220245361,
|
|
"epoch": 3.0971207563386334,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004365946126627699,
|
|
"loss": 5.4189,
|
|
"mean_token_accuracy": 0.2017338365316391,
|
|
"num_tokens": 6551634.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.981328535079956,
|
|
"epoch": 3.101418134937688,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043636308200967433,
|
|
"loss": 5.4241,
|
|
"mean_token_accuracy": 0.2000526711344719,
|
|
"num_tokens": 6560695.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.813827800750732,
|
|
"epoch": 3.1057155135367425,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004361311990041039,
|
|
"loss": 5.3344,
|
|
"mean_token_accuracy": 0.1969393327832222,
|
|
"num_tokens": 6569086.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.89613938331604,
|
|
"epoch": 3.110012892135797,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043589896415239843,
|
|
"loss": 5.4161,
|
|
"mean_token_accuracy": 0.19979367852211,
|
|
"num_tokens": 6578287.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.952451086044311,
|
|
"epoch": 3.1143102707348516,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043566637796166595,
|
|
"loss": 5.4753,
|
|
"mean_token_accuracy": 0.19049297720193864,
|
|
"num_tokens": 6587015.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.962173509597778,
|
|
"epoch": 3.118607649333906,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043543344093978186,
|
|
"loss": 5.5175,
|
|
"mean_token_accuracy": 0.18538623303174973,
|
|
"num_tokens": 6596187.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.912763595581055,
|
|
"epoch": 3.122905027932961,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043520015359538745,
|
|
"loss": 5.3898,
|
|
"mean_token_accuracy": 0.19703881144523622,
|
|
"num_tokens": 6605226.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.858266019821167,
|
|
"epoch": 3.1272024065320156,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004349665164378891,
|
|
"loss": 5.4371,
|
|
"mean_token_accuracy": 0.1935065433382988,
|
|
"num_tokens": 6613232.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.901705503463745,
|
|
"epoch": 3.13149978513107,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043473252997745684,
|
|
"loss": 5.4392,
|
|
"mean_token_accuracy": 0.19095546007156372,
|
|
"num_tokens": 6622247.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.985095024108887,
|
|
"epoch": 3.1357971637301247,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00043449819472502366,
|
|
"loss": 5.3871,
|
|
"mean_token_accuracy": 0.19558012783527373,
|
|
"num_tokens": 6630883.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.880083703994751,
|
|
"epoch": 3.1400945423291793,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004342635111922841,
|
|
"loss": 5.5374,
|
|
"mean_token_accuracy": 0.19031234830617905,
|
|
"num_tokens": 6639399.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.943640947341919,
|
|
"epoch": 3.144391920928234,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004340284798916931,
|
|
"loss": 5.433,
|
|
"mean_token_accuracy": 0.19181231111288072,
|
|
"num_tokens": 6649288.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.898565721511841,
|
|
"epoch": 3.1486892995272884,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004337931013364653,
|
|
"loss": 5.3804,
|
|
"mean_token_accuracy": 0.19686342626810074,
|
|
"num_tokens": 6658670.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.893218088150024,
|
|
"epoch": 3.152986678126343,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000433557376040573,
|
|
"loss": 5.4538,
|
|
"mean_token_accuracy": 0.196715846657753,
|
|
"num_tokens": 6667302.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.956412696838379,
|
|
"epoch": 3.1572840567253975,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043332130451874645,
|
|
"loss": 5.4965,
|
|
"mean_token_accuracy": 0.1952889919281006,
|
|
"num_tokens": 6677393.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.933012199401856,
|
|
"epoch": 3.161581435324452,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043308488728647127,
|
|
"loss": 5.4744,
|
|
"mean_token_accuracy": 0.1893087148666382,
|
|
"num_tokens": 6686727.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.877901268005371,
|
|
"epoch": 3.1658788139235066,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004328481248599882,
|
|
"loss": 5.3869,
|
|
"mean_token_accuracy": 0.19530683755874634,
|
|
"num_tokens": 6696116.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.934775876998901,
|
|
"epoch": 3.170176192522561,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004326110177562918,
|
|
"loss": 5.4945,
|
|
"mean_token_accuracy": 0.18531183749437333,
|
|
"num_tokens": 6704640.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.860188579559326,
|
|
"epoch": 3.1744735711216157,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043237356649312926,
|
|
"loss": 5.3497,
|
|
"mean_token_accuracy": 0.20377567410469055,
|
|
"num_tokens": 6713663.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.891902303695678,
|
|
"epoch": 3.17877094972067,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004321357715889991,
|
|
"loss": 5.4868,
|
|
"mean_token_accuracy": 0.18891167342662812,
|
|
"num_tokens": 6722965.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.922307395935059,
|
|
"epoch": 3.1830683283197247,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004318976335631505,
|
|
"loss": 5.4553,
|
|
"mean_token_accuracy": 0.19657856673002244,
|
|
"num_tokens": 6732776.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.93743257522583,
|
|
"epoch": 3.1873657069187797,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043165915293558155,
|
|
"loss": 5.4328,
|
|
"mean_token_accuracy": 0.19283491969108582,
|
|
"num_tokens": 6741309.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.882253980636596,
|
|
"epoch": 3.1916630855178343,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004314203302270388,
|
|
"loss": 5.4972,
|
|
"mean_token_accuracy": 0.19080058485269547,
|
|
"num_tokens": 6750584.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.9671601295471195,
|
|
"epoch": 3.195960464116889,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004311811659590154,
|
|
"loss": 5.4717,
|
|
"mean_token_accuracy": 0.19037737101316451,
|
|
"num_tokens": 6759344.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.958385229110718,
|
|
"epoch": 3.2002578427159434,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004309416606537507,
|
|
"loss": 5.6122,
|
|
"mean_token_accuracy": 0.18305473029613495,
|
|
"num_tokens": 6770345.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.974073982238769,
|
|
"epoch": 3.204555221314998,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043070181483422843,
|
|
"loss": 5.5015,
|
|
"mean_token_accuracy": 0.18963303416967392,
|
|
"num_tokens": 6779991.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.910816812515259,
|
|
"epoch": 3.2088525999140525,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000430461629024176,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.19442620873451233,
|
|
"num_tokens": 6788972.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.827464151382446,
|
|
"epoch": 3.213149978513107,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004302211037480634,
|
|
"loss": 5.3772,
|
|
"mean_token_accuracy": 0.19249555021524428,
|
|
"num_tokens": 6796967.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.8557556629180905,
|
|
"epoch": 3.2174473571121616,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004299802395311015,
|
|
"loss": 5.4743,
|
|
"mean_token_accuracy": 0.19575096070766448,
|
|
"num_tokens": 6805961.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.86885895729065,
|
|
"epoch": 3.221744735711216,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004297390368992414,
|
|
"loss": 5.3787,
|
|
"mean_token_accuracy": 0.19657269567251207,
|
|
"num_tokens": 6814657.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.923396444320678,
|
|
"epoch": 3.2260421143102707,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042949749637917353,
|
|
"loss": 5.4168,
|
|
"mean_token_accuracy": 0.1941995695233345,
|
|
"num_tokens": 6823095.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.889871215820312,
|
|
"epoch": 3.230339492909325,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004292556184983256,
|
|
"loss": 5.4421,
|
|
"mean_token_accuracy": 0.1958567351102829,
|
|
"num_tokens": 6832195.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.969770717620849,
|
|
"epoch": 3.2346368715083798,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004290134037848623,
|
|
"loss": 5.575,
|
|
"mean_token_accuracy": 0.1806161344051361,
|
|
"num_tokens": 6840922.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.95328722000122,
|
|
"epoch": 3.2389342501074343,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042877085276768386,
|
|
"loss": 5.4178,
|
|
"mean_token_accuracy": 0.20026799887418748,
|
|
"num_tokens": 6849182.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.878392887115479,
|
|
"epoch": 3.243231628706489,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042852796597642455,
|
|
"loss": 5.408,
|
|
"mean_token_accuracy": 0.19837529808282853,
|
|
"num_tokens": 6857932.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.953036594390869,
|
|
"epoch": 3.247529007305544,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004282847439414522,
|
|
"loss": 5.5606,
|
|
"mean_token_accuracy": 0.18480827659368515,
|
|
"num_tokens": 6867283.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.960994386672974,
|
|
"epoch": 3.2518263859045984,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004280411871938664,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.18728075176477432,
|
|
"num_tokens": 6876123.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.969729375839234,
|
|
"epoch": 3.256123764503653,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004277972962654979,
|
|
"loss": 5.4539,
|
|
"mean_token_accuracy": 0.19224015027284622,
|
|
"num_tokens": 6885239.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.8932037353515625,
|
|
"epoch": 3.2604211431027075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004275530716889069,
|
|
"loss": 5.5146,
|
|
"mean_token_accuracy": 0.18382496386766434,
|
|
"num_tokens": 6895061.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.947560358047485,
|
|
"epoch": 3.264718521701762,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004273085139973822,
|
|
"loss": 5.5657,
|
|
"mean_token_accuracy": 0.1781401515007019,
|
|
"num_tokens": 6903828.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.989862442016602,
|
|
"epoch": 3.2690159003008166,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004270636237249401,
|
|
"loss": 5.4777,
|
|
"mean_token_accuracy": 0.18864577561616896,
|
|
"num_tokens": 6912805.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.920726633071899,
|
|
"epoch": 3.273313278899871,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042681840140632314,
|
|
"loss": 5.5243,
|
|
"mean_token_accuracy": 0.18295771330595018,
|
|
"num_tokens": 6922165.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.956135368347168,
|
|
"epoch": 3.2776106574989257,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004265728475769989,
|
|
"loss": 5.4939,
|
|
"mean_token_accuracy": 0.1879052475094795,
|
|
"num_tokens": 6931677.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.918879604339599,
|
|
"epoch": 3.28190803609798,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004263269627731586,
|
|
"loss": 5.452,
|
|
"mean_token_accuracy": 0.192815600335598,
|
|
"num_tokens": 6940486.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.825883960723877,
|
|
"epoch": 3.2862054146970348,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004260807475317164,
|
|
"loss": 5.4745,
|
|
"mean_token_accuracy": 0.18577916026115418,
|
|
"num_tokens": 6948990.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.979638195037841,
|
|
"epoch": 3.2905027932960893,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004258342023903081,
|
|
"loss": 5.5953,
|
|
"mean_token_accuracy": 0.18115273416042327,
|
|
"num_tokens": 6959311.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.957726049423218,
|
|
"epoch": 3.294800171895144,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042558732788728975,
|
|
"loss": 5.3649,
|
|
"mean_token_accuracy": 0.20235307216644288,
|
|
"num_tokens": 6968619.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.867683029174804,
|
|
"epoch": 3.2990975504941984,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042534012456173643,
|
|
"loss": 5.4398,
|
|
"mean_token_accuracy": 0.1914222314953804,
|
|
"num_tokens": 6977469.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.841017484664917,
|
|
"epoch": 3.303394929093253,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042509259295344157,
|
|
"loss": 5.4285,
|
|
"mean_token_accuracy": 0.18821925073862075,
|
|
"num_tokens": 6986772.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.910496807098388,
|
|
"epoch": 3.3076923076923075,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042484473360291514,
|
|
"loss": 5.4393,
|
|
"mean_token_accuracy": 0.19060401618480682,
|
|
"num_tokens": 6993937.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.86693377494812,
|
|
"epoch": 3.311989686291362,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042459654705138294,
|
|
"loss": 5.497,
|
|
"mean_token_accuracy": 0.19289185404777526,
|
|
"num_tokens": 7003222.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.9119508266448975,
|
|
"epoch": 3.316287064890417,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004243480338407853,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.19899186342954636,
|
|
"num_tokens": 7012055.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.892843008041382,
|
|
"epoch": 3.3205844434894716,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004240991945137755,
|
|
"loss": 5.4592,
|
|
"mean_token_accuracy": 0.19213219434022905,
|
|
"num_tokens": 7021036.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.882801103591919,
|
|
"epoch": 3.324881822088526,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042385002961371944,
|
|
"loss": 5.4441,
|
|
"mean_token_accuracy": 0.19504359364509583,
|
|
"num_tokens": 7030450.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.978818750381469,
|
|
"epoch": 3.3291792006875807,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004236005396846935,
|
|
"loss": 5.5439,
|
|
"mean_token_accuracy": 0.1879236653447151,
|
|
"num_tokens": 7039740.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.946993112564087,
|
|
"epoch": 3.333476579286635,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042335072527148406,
|
|
"loss": 5.5256,
|
|
"mean_token_accuracy": 0.19050987511873246,
|
|
"num_tokens": 7050430.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.85164065361023,
|
|
"epoch": 3.3377739578856898,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004231005869195859,
|
|
"loss": 5.5069,
|
|
"mean_token_accuracy": 0.18632889091968535,
|
|
"num_tokens": 7059477.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.922405767440796,
|
|
"epoch": 3.3420713364847443,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004228501251752011,
|
|
"loss": 5.4601,
|
|
"mean_token_accuracy": 0.1952619045972824,
|
|
"num_tokens": 7067805.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.881201887130738,
|
|
"epoch": 3.346368715083799,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042259934058523814,
|
|
"loss": 5.46,
|
|
"mean_token_accuracy": 0.18905477821826935,
|
|
"num_tokens": 7077606.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.918180227279663,
|
|
"epoch": 3.3506660936828534,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042234823369731027,
|
|
"loss": 5.4043,
|
|
"mean_token_accuracy": 0.1953267216682434,
|
|
"num_tokens": 7085647.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.835781908035278,
|
|
"epoch": 3.354963472281908,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042209680505973465,
|
|
"loss": 5.449,
|
|
"mean_token_accuracy": 0.1939581647515297,
|
|
"num_tokens": 7095298.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.841428470611572,
|
|
"epoch": 3.3592608508809625,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004218450552215308,
|
|
"loss": 5.5157,
|
|
"mean_token_accuracy": 0.194383442401886,
|
|
"num_tokens": 7105207.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.941448974609375,
|
|
"epoch": 3.363558229480017,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004215929847324199,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.18115754574537277,
|
|
"num_tokens": 7114833.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.963439035415649,
|
|
"epoch": 3.3678556080790716,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000421340594142823,
|
|
"loss": 5.3787,
|
|
"mean_token_accuracy": 0.19882705360651015,
|
|
"num_tokens": 7123608.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.8740808963775635,
|
|
"epoch": 3.3721529866781266,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042108788400386035,
|
|
"loss": 5.4499,
|
|
"mean_token_accuracy": 0.19346580952405928,
|
|
"num_tokens": 7132250.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.880605030059814,
|
|
"epoch": 3.376450365277181,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004208348548673498,
|
|
"loss": 5.5399,
|
|
"mean_token_accuracy": 0.19135694503784179,
|
|
"num_tokens": 7142086.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.951083946228027,
|
|
"epoch": 3.3807477438762357,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000420581507285806,
|
|
"loss": 5.4858,
|
|
"mean_token_accuracy": 0.1825383946299553,
|
|
"num_tokens": 7152434.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.82304835319519,
|
|
"epoch": 3.38504512247529,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004203278418124386,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.19977713227272034,
|
|
"num_tokens": 7163041.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.829355192184448,
|
|
"epoch": 3.3893425010743448,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004200738590011518,
|
|
"loss": 5.4173,
|
|
"mean_token_accuracy": 0.19818853884935378,
|
|
"num_tokens": 7171875.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.888874340057373,
|
|
"epoch": 3.3936398796733993,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041981955940654245,
|
|
"loss": 5.5242,
|
|
"mean_token_accuracy": 0.1952082931995392,
|
|
"num_tokens": 7180803.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.918650579452515,
|
|
"epoch": 3.397937258272454,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004195649435838992,
|
|
"loss": 5.5527,
|
|
"mean_token_accuracy": 0.17839486598968507,
|
|
"num_tokens": 7190661.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.824749708175659,
|
|
"epoch": 3.4022346368715084,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004193100120892013,
|
|
"loss": 5.3825,
|
|
"mean_token_accuracy": 0.20471955984830856,
|
|
"num_tokens": 7199357.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.87763524055481,
|
|
"epoch": 3.406532015470563,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004190547654791172,
|
|
"loss": 5.5507,
|
|
"mean_token_accuracy": 0.18516142815351486,
|
|
"num_tokens": 7209856.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.960424184799194,
|
|
"epoch": 3.4108293940696175,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041879920431100347,
|
|
"loss": 5.5182,
|
|
"mean_token_accuracy": 0.18072724491357803,
|
|
"num_tokens": 7218778.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.929759693145752,
|
|
"epoch": 3.415126772668672,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004185433291429036,
|
|
"loss": 5.5383,
|
|
"mean_token_accuracy": 0.19149455428123474,
|
|
"num_tokens": 7228442.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.9323328018188475,
|
|
"epoch": 3.4194241512677266,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041828714053354665,
|
|
"loss": 5.5232,
|
|
"mean_token_accuracy": 0.18421948850154876,
|
|
"num_tokens": 7238724.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.825714445114135,
|
|
"epoch": 3.423721529866781,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004180306390423462,
|
|
"loss": 5.48,
|
|
"mean_token_accuracy": 0.19667282402515412,
|
|
"num_tokens": 7247844.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.873914575576782,
|
|
"epoch": 3.4280189084658357,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041777382522939884,
|
|
"loss": 5.5471,
|
|
"mean_token_accuracy": 0.18860624134540557,
|
|
"num_tokens": 7257260.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.9357811450958256,
|
|
"epoch": 3.4323162870648902,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041751669965548344,
|
|
"loss": 5.5448,
|
|
"mean_token_accuracy": 0.18522292822599412,
|
|
"num_tokens": 7266890.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.9294140338897705,
|
|
"epoch": 3.4366136656639448,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041725926288205945,
|
|
"loss": 5.5664,
|
|
"mean_token_accuracy": 0.18018038868904113,
|
|
"num_tokens": 7276114.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.4366136656639448,
|
|
"eval_entropy": 5.70672602589066,
|
|
"eval_loss": 5.984828472137451,
|
|
"eval_mean_token_accuracy": 0.17211216785483532,
|
|
"eval_num_tokens": 7276114.0,
|
|
"eval_runtime": 2.0584,
|
|
"eval_samples_per_second": 1724.118,
|
|
"eval_steps_per_second": 215.697,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"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": 1640861880145920.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|