12167 lines
332 KiB
JSON
12167 lines
332 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 5.964694181999006,
|
|
"eval_steps": 500,
|
|
"global_step": 6000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.69195852279663,
|
|
"epoch": 0.004972650422675286,
|
|
"grad_norm": 13.0,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8154,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9089.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69191370010376,
|
|
"epoch": 0.009945300845350571,
|
|
"grad_norm": 12.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.8258,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 21650.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.69098253250122,
|
|
"epoch": 0.014917951268025857,
|
|
"grad_norm": 9.625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5784,
|
|
"mean_token_accuracy": 0.020547608099877834,
|
|
"num_tokens": 31067.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.673026275634765,
|
|
"epoch": 0.019890601690701143,
|
|
"grad_norm": 6.40625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3334,
|
|
"mean_token_accuracy": 0.041073380038142204,
|
|
"num_tokens": 40816.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.540556049346923,
|
|
"epoch": 0.02486325211337643,
|
|
"grad_norm": 4.28125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.1325,
|
|
"mean_token_accuracy": 0.0424393804743886,
|
|
"num_tokens": 51065.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.523736000061035,
|
|
"epoch": 0.029835902536051714,
|
|
"grad_norm": 3.359375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9451,
|
|
"mean_token_accuracy": 0.044780130684375766,
|
|
"num_tokens": 60539.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.482940578460694,
|
|
"epoch": 0.034808552958727,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9267,
|
|
"mean_token_accuracy": 0.04583238400518894,
|
|
"num_tokens": 71010.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.497202396392822,
|
|
"epoch": 0.039781203381402286,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.8766,
|
|
"mean_token_accuracy": 0.04677955023944378,
|
|
"num_tokens": 81394.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.495645523071289,
|
|
"epoch": 0.04475385380407757,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.9242,
|
|
"mean_token_accuracy": 0.04332513865083456,
|
|
"num_tokens": 92491.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.476744461059571,
|
|
"epoch": 0.04972650422675286,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.8095,
|
|
"mean_token_accuracy": 0.045694665610790254,
|
|
"num_tokens": 102113.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.468534755706788,
|
|
"epoch": 0.05469915464942814,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8159,
|
|
"mean_token_accuracy": 0.051046956703066824,
|
|
"num_tokens": 112211.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.431697177886964,
|
|
"epoch": 0.05967180507210343,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7102,
|
|
"mean_token_accuracy": 0.05394814908504486,
|
|
"num_tokens": 122166.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.248080635070801,
|
|
"epoch": 0.06464445549477872,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.6181,
|
|
"mean_token_accuracy": 0.059726152941584586,
|
|
"num_tokens": 131016.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.313014602661132,
|
|
"epoch": 0.069617105917454,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.587,
|
|
"mean_token_accuracy": 0.06370560713112354,
|
|
"num_tokens": 141369.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.329190254211426,
|
|
"epoch": 0.07458975634012929,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5355,
|
|
"mean_token_accuracy": 0.058842772617936134,
|
|
"num_tokens": 150939.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.338141822814942,
|
|
"epoch": 0.07956240676280457,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.478,
|
|
"mean_token_accuracy": 0.05786029174923897,
|
|
"num_tokens": 159981.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.291361045837402,
|
|
"epoch": 0.08453505718547986,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.4043,
|
|
"mean_token_accuracy": 0.05959840677678585,
|
|
"num_tokens": 170139.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.275171947479247,
|
|
"epoch": 0.08950770760815514,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3294,
|
|
"mean_token_accuracy": 0.06485622599720955,
|
|
"num_tokens": 180746.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.15320930480957,
|
|
"epoch": 0.09448035803083044,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.1862,
|
|
"mean_token_accuracy": 0.06401227861642837,
|
|
"num_tokens": 190250.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.223065280914307,
|
|
"epoch": 0.09945300845350571,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2406,
|
|
"mean_token_accuracy": 0.061084074527025224,
|
|
"num_tokens": 201185.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.13264331817627,
|
|
"epoch": 0.10442565887618101,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.16,
|
|
"mean_token_accuracy": 0.05885303355753422,
|
|
"num_tokens": 211476.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.12833375930786,
|
|
"epoch": 0.10939830929885629,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0533,
|
|
"mean_token_accuracy": 0.062058281153440475,
|
|
"num_tokens": 220965.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.080890655517578,
|
|
"epoch": 0.11437095972153158,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9846,
|
|
"mean_token_accuracy": 0.05859192833304405,
|
|
"num_tokens": 231539.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.94415922164917,
|
|
"epoch": 0.11934361014420686,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.881,
|
|
"mean_token_accuracy": 0.06372438780963421,
|
|
"num_tokens": 241206.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.881352806091309,
|
|
"epoch": 0.12431626056688215,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7735,
|
|
"mean_token_accuracy": 0.06269489601254463,
|
|
"num_tokens": 250629.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.732184791564942,
|
|
"epoch": 0.12928891098955744,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.7496,
|
|
"mean_token_accuracy": 0.05945408381521702,
|
|
"num_tokens": 260705.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.715274715423584,
|
|
"epoch": 0.13426156141223272,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6842,
|
|
"mean_token_accuracy": 0.05957487896084786,
|
|
"num_tokens": 270499.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.532818031311034,
|
|
"epoch": 0.139234211834908,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6647,
|
|
"mean_token_accuracy": 0.05812281183898449,
|
|
"num_tokens": 279998.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.431089496612548,
|
|
"epoch": 0.14420686225758328,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.5342,
|
|
"mean_token_accuracy": 0.06715737730264663,
|
|
"num_tokens": 290322.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.28282127380371,
|
|
"epoch": 0.14917951268025859,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.4927,
|
|
"mean_token_accuracy": 0.06783988289535045,
|
|
"num_tokens": 299893.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.202215671539307,
|
|
"epoch": 0.15415216310293386,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.4742,
|
|
"mean_token_accuracy": 0.06662740968167782,
|
|
"num_tokens": 309191.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.073563289642333,
|
|
"epoch": 0.15912481352560914,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.4087,
|
|
"mean_token_accuracy": 0.06572333909571171,
|
|
"num_tokens": 318494.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.005878067016601,
|
|
"epoch": 0.16409746394828442,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.3906,
|
|
"mean_token_accuracy": 0.06560419239103794,
|
|
"num_tokens": 328299.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.962747955322266,
|
|
"epoch": 0.16907011437095973,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4667,
|
|
"mean_token_accuracy": 0.06214671768248081,
|
|
"num_tokens": 338878.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.892442321777343,
|
|
"epoch": 0.174042764793635,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.3855,
|
|
"mean_token_accuracy": 0.06676384881138801,
|
|
"num_tokens": 348508.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.82924509048462,
|
|
"epoch": 0.1790154152163103,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.3408,
|
|
"mean_token_accuracy": 0.06486317589879036,
|
|
"num_tokens": 358233.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.784028911590577,
|
|
"epoch": 0.1839880656389856,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.2883,
|
|
"mean_token_accuracy": 0.06572877354919911,
|
|
"num_tokens": 367297.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.75277099609375,
|
|
"epoch": 0.18896071606166087,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.2958,
|
|
"mean_token_accuracy": 0.06614930555224419,
|
|
"num_tokens": 376880.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.710809421539306,
|
|
"epoch": 0.19393336648433615,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.2937,
|
|
"mean_token_accuracy": 0.06907838135957718,
|
|
"num_tokens": 386806.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.63250789642334,
|
|
"epoch": 0.19890601690701143,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.257,
|
|
"mean_token_accuracy": 0.07009859867393971,
|
|
"num_tokens": 396161.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.667188930511475,
|
|
"epoch": 0.20387866732968674,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.2966,
|
|
"mean_token_accuracy": 0.0686015360057354,
|
|
"num_tokens": 405643.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.586182785034179,
|
|
"epoch": 0.20885131775236201,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.249,
|
|
"mean_token_accuracy": 0.07686658278107643,
|
|
"num_tokens": 414367.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.671557807922364,
|
|
"epoch": 0.2138239681750373,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.3009,
|
|
"mean_token_accuracy": 0.07021823003888131,
|
|
"num_tokens": 424430.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.60766077041626,
|
|
"epoch": 0.21879661859771257,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.2723,
|
|
"mean_token_accuracy": 0.07040721997618675,
|
|
"num_tokens": 435176.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.483222579956054,
|
|
"epoch": 0.22376926902038788,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.2963,
|
|
"mean_token_accuracy": 0.07160340994596481,
|
|
"num_tokens": 445611.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.615241146087646,
|
|
"epoch": 0.22874191944306316,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.2193,
|
|
"mean_token_accuracy": 0.07408608645200729,
|
|
"num_tokens": 455143.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.600674724578857,
|
|
"epoch": 0.23371456986573844,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.2898,
|
|
"mean_token_accuracy": 0.07503409050405026,
|
|
"num_tokens": 465507.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.539074230194093,
|
|
"epoch": 0.23868722028841372,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.3078,
|
|
"mean_token_accuracy": 0.0719807006418705,
|
|
"num_tokens": 476152.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.533562564849854,
|
|
"epoch": 0.24365987071108902,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.1356,
|
|
"mean_token_accuracy": 0.07810752168297767,
|
|
"num_tokens": 486341.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.470274734497071,
|
|
"epoch": 0.2486325211337643,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.0909,
|
|
"mean_token_accuracy": 0.07914422042667865,
|
|
"num_tokens": 495915.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.434992504119872,
|
|
"epoch": 0.2536051715564396,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.1795,
|
|
"mean_token_accuracy": 0.07838663607835769,
|
|
"num_tokens": 505666.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.52521858215332,
|
|
"epoch": 0.2585778219791149,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.0848,
|
|
"mean_token_accuracy": 0.07590681686997414,
|
|
"num_tokens": 515411.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.440220355987549,
|
|
"epoch": 0.26355047240179014,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.1808,
|
|
"mean_token_accuracy": 0.07005526758730411,
|
|
"num_tokens": 525018.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.485704612731933,
|
|
"epoch": 0.26852312282446544,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.132,
|
|
"mean_token_accuracy": 0.07571323961019516,
|
|
"num_tokens": 535056.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.444093799591064,
|
|
"epoch": 0.27349577324714075,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.0943,
|
|
"mean_token_accuracy": 0.07863550931215287,
|
|
"num_tokens": 545115.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.46326379776001,
|
|
"epoch": 0.278468423669816,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.1362,
|
|
"mean_token_accuracy": 0.07623238041996956,
|
|
"num_tokens": 554429.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.404027462005615,
|
|
"epoch": 0.2834410740924913,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.0743,
|
|
"mean_token_accuracy": 0.0847582258284092,
|
|
"num_tokens": 563357.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.44946584701538,
|
|
"epoch": 0.28841372451516656,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.114,
|
|
"mean_token_accuracy": 0.07873134687542915,
|
|
"num_tokens": 574480.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.350453853607178,
|
|
"epoch": 0.29338637493784187,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.1023,
|
|
"mean_token_accuracy": 0.07694951295852662,
|
|
"num_tokens": 583722.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.420344829559326,
|
|
"epoch": 0.29835902536051717,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.0479,
|
|
"mean_token_accuracy": 0.08190468847751617,
|
|
"num_tokens": 593644.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.371365547180176,
|
|
"epoch": 0.3033316757831924,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.1017,
|
|
"mean_token_accuracy": 0.07948060184717179,
|
|
"num_tokens": 603547.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.302706527709962,
|
|
"epoch": 0.30830432620586773,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.1257,
|
|
"mean_token_accuracy": 0.07789697460830211,
|
|
"num_tokens": 613880.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.415989398956299,
|
|
"epoch": 0.31327697662854304,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.0974,
|
|
"mean_token_accuracy": 0.07658020406961441,
|
|
"num_tokens": 624213.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.416389560699463,
|
|
"epoch": 0.3182496270512183,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.128,
|
|
"mean_token_accuracy": 0.08172445297241211,
|
|
"num_tokens": 634350.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.42117624282837,
|
|
"epoch": 0.3232222774738936,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.1272,
|
|
"mean_token_accuracy": 0.08092540055513382,
|
|
"num_tokens": 644040.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.33439235687256,
|
|
"epoch": 0.32819492789656884,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.0773,
|
|
"mean_token_accuracy": 0.0839149996638298,
|
|
"num_tokens": 652915.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.363460159301757,
|
|
"epoch": 0.33316757831924415,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.0669,
|
|
"mean_token_accuracy": 0.07783392369747162,
|
|
"num_tokens": 662665.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.326759719848633,
|
|
"epoch": 0.33814022874191946,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.0773,
|
|
"mean_token_accuracy": 0.08024705946445465,
|
|
"num_tokens": 671667.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.384268760681152,
|
|
"epoch": 0.3431128791645947,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.0868,
|
|
"mean_token_accuracy": 0.0760427363216877,
|
|
"num_tokens": 680432.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.345528602600098,
|
|
"epoch": 0.34808552958727,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.0811,
|
|
"mean_token_accuracy": 0.08016384020447731,
|
|
"num_tokens": 690184.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.333233165740968,
|
|
"epoch": 0.3530581800099453,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.9808,
|
|
"mean_token_accuracy": 0.08433079347014427,
|
|
"num_tokens": 699147.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.210507869720459,
|
|
"epoch": 0.3580308304326206,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.9962,
|
|
"mean_token_accuracy": 0.08303841575980186,
|
|
"num_tokens": 708696.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.333552265167237,
|
|
"epoch": 0.3630034808552959,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.9812,
|
|
"mean_token_accuracy": 0.08292912170290948,
|
|
"num_tokens": 718438.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.339497661590576,
|
|
"epoch": 0.3679761312779712,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.1143,
|
|
"mean_token_accuracy": 0.07707406133413315,
|
|
"num_tokens": 728157.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.221984767913819,
|
|
"epoch": 0.37294878170064644,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.0113,
|
|
"mean_token_accuracy": 0.083165742456913,
|
|
"num_tokens": 737007.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.301848316192627,
|
|
"epoch": 0.37792143212332174,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.9361,
|
|
"mean_token_accuracy": 0.08524221666157246,
|
|
"num_tokens": 746873.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.174521112442017,
|
|
"epoch": 0.382894082545997,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.856,
|
|
"mean_token_accuracy": 0.08981255739927292,
|
|
"num_tokens": 755486.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.252612590789795,
|
|
"epoch": 0.3878667329686723,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.9822,
|
|
"mean_token_accuracy": 0.07825585678219796,
|
|
"num_tokens": 765481.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.236901569366456,
|
|
"epoch": 0.3928393833913476,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.937,
|
|
"mean_token_accuracy": 0.08628773540258408,
|
|
"num_tokens": 774335.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.332719326019287,
|
|
"epoch": 0.39781203381402286,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.0001,
|
|
"mean_token_accuracy": 0.08222218602895737,
|
|
"num_tokens": 784307.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.230411338806153,
|
|
"epoch": 0.40278468423669817,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.0406,
|
|
"mean_token_accuracy": 0.07930415198206901,
|
|
"num_tokens": 795058.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.313278293609619,
|
|
"epoch": 0.40775733465937347,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.9507,
|
|
"mean_token_accuracy": 0.08477597758173942,
|
|
"num_tokens": 805687.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.19376473426819,
|
|
"epoch": 0.4127299850820487,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.989,
|
|
"mean_token_accuracy": 0.08413419276475906,
|
|
"num_tokens": 815138.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.226968097686768,
|
|
"epoch": 0.41770263550472403,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9451,
|
|
"mean_token_accuracy": 0.08170225620269775,
|
|
"num_tokens": 825009.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.161935615539551,
|
|
"epoch": 0.4226752859273993,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.9202,
|
|
"mean_token_accuracy": 0.0900972880423069,
|
|
"num_tokens": 834708.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.352538585662842,
|
|
"epoch": 0.4276479363500746,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.9872,
|
|
"mean_token_accuracy": 0.08343984819948673,
|
|
"num_tokens": 843888.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.144132709503173,
|
|
"epoch": 0.4326205867727499,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.9481,
|
|
"mean_token_accuracy": 0.08321668803691865,
|
|
"num_tokens": 853057.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.240882396697998,
|
|
"epoch": 0.43759323719542514,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.0054,
|
|
"mean_token_accuracy": 0.07688518613576889,
|
|
"num_tokens": 862701.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.124331283569337,
|
|
"epoch": 0.44256588761810045,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.8978,
|
|
"mean_token_accuracy": 0.0850852258503437,
|
|
"num_tokens": 872129.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.162071371078492,
|
|
"epoch": 0.44753853804077576,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.9163,
|
|
"mean_token_accuracy": 0.08625531122088433,
|
|
"num_tokens": 883166.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.160173559188843,
|
|
"epoch": 0.452511188463451,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.7103,
|
|
"mean_token_accuracy": 0.10269899219274521,
|
|
"num_tokens": 894159.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.14470615386963,
|
|
"epoch": 0.4574838388861263,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.9094,
|
|
"mean_token_accuracy": 0.08693445995450019,
|
|
"num_tokens": 904588.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.188994216918946,
|
|
"epoch": 0.46245648930880157,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.8764,
|
|
"mean_token_accuracy": 0.08472439050674438,
|
|
"num_tokens": 915554.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.080974769592284,
|
|
"epoch": 0.4674291397314769,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.9153,
|
|
"mean_token_accuracy": 0.08615779727697373,
|
|
"num_tokens": 925546.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.106526470184326,
|
|
"epoch": 0.4724017901541522,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.8827,
|
|
"mean_token_accuracy": 0.08863506764173508,
|
|
"num_tokens": 936615.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.171993637084961,
|
|
"epoch": 0.47737444057682743,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.9148,
|
|
"mean_token_accuracy": 0.08405414745211601,
|
|
"num_tokens": 947374.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.07227544784546,
|
|
"epoch": 0.48234709099950274,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.8263,
|
|
"mean_token_accuracy": 0.0881466455757618,
|
|
"num_tokens": 957772.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.149549674987792,
|
|
"epoch": 0.48731974142217804,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.9265,
|
|
"mean_token_accuracy": 0.08613847866654396,
|
|
"num_tokens": 967724.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.014617347717286,
|
|
"epoch": 0.4922923918448533,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.821,
|
|
"mean_token_accuracy": 0.08873307779431343,
|
|
"num_tokens": 976497.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.061779451370239,
|
|
"epoch": 0.4972650422675286,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.8061,
|
|
"mean_token_accuracy": 0.08952893353998662,
|
|
"num_tokens": 986207.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4972650422675286,
|
|
"eval_entropy": 8.12199338789909,
|
|
"eval_loss": 7.925317764282227,
|
|
"eval_mean_token_accuracy": 0.0847069110511051,
|
|
"eval_num_tokens": 986207.0,
|
|
"eval_runtime": 11.3595,
|
|
"eval_samples_per_second": 305.031,
|
|
"eval_steps_per_second": 38.206,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.053079652786256,
|
|
"epoch": 0.5022376926902039,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.779,
|
|
"mean_token_accuracy": 0.09456762075424194,
|
|
"num_tokens": 995568.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.076926040649415,
|
|
"epoch": 0.5072103431128792,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.8379,
|
|
"mean_token_accuracy": 0.08805601447820663,
|
|
"num_tokens": 1005829.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.036222219467163,
|
|
"epoch": 0.5121829935355544,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.7909,
|
|
"mean_token_accuracy": 0.08699572384357453,
|
|
"num_tokens": 1015290.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.061750364303588,
|
|
"epoch": 0.5171556439582298,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.7536,
|
|
"mean_token_accuracy": 0.09390853121876716,
|
|
"num_tokens": 1024479.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.9042600154876705,
|
|
"epoch": 0.522128294380905,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.6663,
|
|
"mean_token_accuracy": 0.09484521970152855,
|
|
"num_tokens": 1034092.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.166246604919433,
|
|
"epoch": 0.5271009448035803,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.9006,
|
|
"mean_token_accuracy": 0.09057568311691284,
|
|
"num_tokens": 1043844.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.903245353698731,
|
|
"epoch": 0.5320735952262556,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.7407,
|
|
"mean_token_accuracy": 0.09305151179432869,
|
|
"num_tokens": 1052646.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.107067775726318,
|
|
"epoch": 0.5370462456489309,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.7945,
|
|
"mean_token_accuracy": 0.08951174244284629,
|
|
"num_tokens": 1062345.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.024613332748412,
|
|
"epoch": 0.5420188960716061,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.7886,
|
|
"mean_token_accuracy": 0.08819786384701729,
|
|
"num_tokens": 1071116.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.918569803237915,
|
|
"epoch": 0.5469915464942815,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.7358,
|
|
"mean_token_accuracy": 0.09049084335565567,
|
|
"num_tokens": 1080562.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.018861722946166,
|
|
"epoch": 0.5519641969169568,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.7855,
|
|
"mean_token_accuracy": 0.08982611894607544,
|
|
"num_tokens": 1091944.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.988247442245483,
|
|
"epoch": 0.556936847339632,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.7154,
|
|
"mean_token_accuracy": 0.09307126253843308,
|
|
"num_tokens": 1101412.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.006781101226807,
|
|
"epoch": 0.5619094977623074,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.8081,
|
|
"mean_token_accuracy": 0.09186437204480172,
|
|
"num_tokens": 1110652.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.9736639022827145,
|
|
"epoch": 0.5668821481849826,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.7146,
|
|
"mean_token_accuracy": 0.09523341283202172,
|
|
"num_tokens": 1120420.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.896536588668823,
|
|
"epoch": 0.5718547986076579,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.775,
|
|
"mean_token_accuracy": 0.08853160478174686,
|
|
"num_tokens": 1130046.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.949215126037598,
|
|
"epoch": 0.5768274490303331,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.7521,
|
|
"mean_token_accuracy": 0.09078693836927414,
|
|
"num_tokens": 1138946.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.021563529968262,
|
|
"epoch": 0.5818000994530085,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.4967,
|
|
"mean_token_accuracy": 0.10361211746931076,
|
|
"num_tokens": 1149440.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.954723596572876,
|
|
"epoch": 0.5867727498756837,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.7857,
|
|
"mean_token_accuracy": 0.08751340731978416,
|
|
"num_tokens": 1158636.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.810801792144775,
|
|
"epoch": 0.591745400298359,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.6241,
|
|
"mean_token_accuracy": 0.0969457134604454,
|
|
"num_tokens": 1167776.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.93477840423584,
|
|
"epoch": 0.5967180507210343,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.6271,
|
|
"mean_token_accuracy": 0.0945424608886242,
|
|
"num_tokens": 1177919.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.868533611297607,
|
|
"epoch": 0.6016907011437096,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.695,
|
|
"mean_token_accuracy": 0.10151977390050888,
|
|
"num_tokens": 1187363.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.040846824645996,
|
|
"epoch": 0.6066633515663848,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.8651,
|
|
"mean_token_accuracy": 0.08632402569055557,
|
|
"num_tokens": 1197329.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.835066843032837,
|
|
"epoch": 0.6116360019890602,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.7643,
|
|
"mean_token_accuracy": 0.08976817056536675,
|
|
"num_tokens": 1207408.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.966505718231201,
|
|
"epoch": 0.6166086524117355,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7505,
|
|
"mean_token_accuracy": 0.08791251480579376,
|
|
"num_tokens": 1219258.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.945406055450439,
|
|
"epoch": 0.6215813028344107,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.7417,
|
|
"mean_token_accuracy": 0.08997605070471763,
|
|
"num_tokens": 1228561.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.796209669113159,
|
|
"epoch": 0.6265539532570861,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.6714,
|
|
"mean_token_accuracy": 0.09316384196281433,
|
|
"num_tokens": 1239665.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.944826412200928,
|
|
"epoch": 0.6315266036797613,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.6703,
|
|
"mean_token_accuracy": 0.09130050241947174,
|
|
"num_tokens": 1249490.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.864301347732544,
|
|
"epoch": 0.6364992541024366,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.6962,
|
|
"mean_token_accuracy": 0.09296346306800843,
|
|
"num_tokens": 1259328.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.925337028503418,
|
|
"epoch": 0.6414719045251119,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.6193,
|
|
"mean_token_accuracy": 0.0926283597946167,
|
|
"num_tokens": 1269523.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.766702365875244,
|
|
"epoch": 0.6464445549477872,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7288,
|
|
"mean_token_accuracy": 0.08978976607322693,
|
|
"num_tokens": 1277961.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.95000376701355,
|
|
"epoch": 0.6514172053704624,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.6807,
|
|
"mean_token_accuracy": 0.09115450382232666,
|
|
"num_tokens": 1288228.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.832488107681274,
|
|
"epoch": 0.6563898557931377,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.6293,
|
|
"mean_token_accuracy": 0.09410025104880333,
|
|
"num_tokens": 1299636.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.785907840728759,
|
|
"epoch": 0.661362506215813,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.6553,
|
|
"mean_token_accuracy": 0.09513049125671387,
|
|
"num_tokens": 1310035.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.74755220413208,
|
|
"epoch": 0.6663351566384883,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.6673,
|
|
"mean_token_accuracy": 0.09929112568497658,
|
|
"num_tokens": 1320751.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.734870958328247,
|
|
"epoch": 0.6713078070611636,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.5335,
|
|
"mean_token_accuracy": 0.09909648820757866,
|
|
"num_tokens": 1331023.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.7550742626190186,
|
|
"epoch": 0.6762804574838389,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.6342,
|
|
"mean_token_accuracy": 0.0891692079603672,
|
|
"num_tokens": 1341814.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.792105770111084,
|
|
"epoch": 0.6812531079065142,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.5512,
|
|
"mean_token_accuracy": 0.09285351634025574,
|
|
"num_tokens": 1351026.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.822172403335571,
|
|
"epoch": 0.6862257583291894,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7478,
|
|
"mean_token_accuracy": 0.08240504153072833,
|
|
"num_tokens": 1361004.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.7296113014221195,
|
|
"epoch": 0.6911984087518648,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.6089,
|
|
"mean_token_accuracy": 0.09175579845905305,
|
|
"num_tokens": 1370729.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.822756958007813,
|
|
"epoch": 0.69617105917454,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.5643,
|
|
"mean_token_accuracy": 0.09511974975466728,
|
|
"num_tokens": 1379525.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.708429861068725,
|
|
"epoch": 0.7011437095972153,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.6297,
|
|
"mean_token_accuracy": 0.09320607557892799,
|
|
"num_tokens": 1389863.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.726954174041748,
|
|
"epoch": 0.7061163600198906,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7481,
|
|
"mean_token_accuracy": 0.08981630355119705,
|
|
"num_tokens": 1399468.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.754483079910278,
|
|
"epoch": 0.7110890104425659,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.5528,
|
|
"mean_token_accuracy": 0.09678518697619438,
|
|
"num_tokens": 1408785.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.69726414680481,
|
|
"epoch": 0.7160616608652411,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.5883,
|
|
"mean_token_accuracy": 0.09907481074333191,
|
|
"num_tokens": 1418666.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.866272592544556,
|
|
"epoch": 0.7210343112879165,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.7005,
|
|
"mean_token_accuracy": 0.09374464005231857,
|
|
"num_tokens": 1428699.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.772790002822876,
|
|
"epoch": 0.7260069617105918,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.5955,
|
|
"mean_token_accuracy": 0.09428214207291603,
|
|
"num_tokens": 1437354.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.768170261383057,
|
|
"epoch": 0.730979612133267,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.5997,
|
|
"mean_token_accuracy": 0.09616516754031182,
|
|
"num_tokens": 1447327.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.792877292633056,
|
|
"epoch": 0.7359522625559424,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6569,
|
|
"mean_token_accuracy": 0.09555381312966346,
|
|
"num_tokens": 1457793.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.718319368362427,
|
|
"epoch": 0.7409249129786176,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.5646,
|
|
"mean_token_accuracy": 0.09642104059457779,
|
|
"num_tokens": 1467143.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.69613127708435,
|
|
"epoch": 0.7458975634012929,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6171,
|
|
"mean_token_accuracy": 0.09400601610541344,
|
|
"num_tokens": 1476422.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.6746134757995605,
|
|
"epoch": 0.7508702138239681,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.55,
|
|
"mean_token_accuracy": 0.10190361589193345,
|
|
"num_tokens": 1486253.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.683488607406616,
|
|
"epoch": 0.7558428642466435,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.5676,
|
|
"mean_token_accuracy": 0.0934619203209877,
|
|
"num_tokens": 1496080.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.70647873878479,
|
|
"epoch": 0.7608155146693187,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.5191,
|
|
"mean_token_accuracy": 0.09906428828835487,
|
|
"num_tokens": 1505532.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.636048603057861,
|
|
"epoch": 0.765788165091994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.4547,
|
|
"mean_token_accuracy": 0.09559990018606186,
|
|
"num_tokens": 1515705.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.626402521133423,
|
|
"epoch": 0.7707608155146694,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.515,
|
|
"mean_token_accuracy": 0.09692738726735114,
|
|
"num_tokens": 1525651.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.607789754867554,
|
|
"epoch": 0.7757334659373446,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5051,
|
|
"mean_token_accuracy": 0.09369031935930253,
|
|
"num_tokens": 1534576.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.578536558151245,
|
|
"epoch": 0.7807061163600199,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.3647,
|
|
"mean_token_accuracy": 0.10836331695318221,
|
|
"num_tokens": 1544765.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.696880531311035,
|
|
"epoch": 0.7856787667826952,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.4927,
|
|
"mean_token_accuracy": 0.10022924914956093,
|
|
"num_tokens": 1553636.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.581558704376221,
|
|
"epoch": 0.7906514172053705,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.4583,
|
|
"mean_token_accuracy": 0.09628499075770378,
|
|
"num_tokens": 1563356.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.529683589935303,
|
|
"epoch": 0.7956240676280457,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.4794,
|
|
"mean_token_accuracy": 0.10385911986231804,
|
|
"num_tokens": 1572508.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.677089595794678,
|
|
"epoch": 0.8005967180507211,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5022,
|
|
"mean_token_accuracy": 0.09889172688126564,
|
|
"num_tokens": 1581583.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.620496559143066,
|
|
"epoch": 0.8055693684733963,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5028,
|
|
"mean_token_accuracy": 0.0993174895644188,
|
|
"num_tokens": 1592114.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.58529052734375,
|
|
"epoch": 0.8105420188960716,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.414,
|
|
"mean_token_accuracy": 0.10554423704743385,
|
|
"num_tokens": 1602417.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.617566537857056,
|
|
"epoch": 0.8155146693187469,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5095,
|
|
"mean_token_accuracy": 0.09752819314599037,
|
|
"num_tokens": 1611753.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.640080833435059,
|
|
"epoch": 0.8204873197414222,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.565,
|
|
"mean_token_accuracy": 0.09824235141277313,
|
|
"num_tokens": 1621959.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.517613458633423,
|
|
"epoch": 0.8254599701640974,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.4706,
|
|
"mean_token_accuracy": 0.09650598615407943,
|
|
"num_tokens": 1630909.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.668545198440552,
|
|
"epoch": 0.8304326205867727,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.4442,
|
|
"mean_token_accuracy": 0.09847426488995552,
|
|
"num_tokens": 1641316.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.467155838012696,
|
|
"epoch": 0.8354052710094481,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.3595,
|
|
"mean_token_accuracy": 0.10589170381426811,
|
|
"num_tokens": 1651506.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.445871210098266,
|
|
"epoch": 0.8403779214321233,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.3611,
|
|
"mean_token_accuracy": 0.1002371571958065,
|
|
"num_tokens": 1659871.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.649908971786499,
|
|
"epoch": 0.8453505718547986,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.578,
|
|
"mean_token_accuracy": 0.09238532930612564,
|
|
"num_tokens": 1670178.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.569011211395264,
|
|
"epoch": 0.8503232222774739,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.4567,
|
|
"mean_token_accuracy": 0.09851670637726784,
|
|
"num_tokens": 1680131.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.358003282546997,
|
|
"epoch": 0.8552958727001492,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.3453,
|
|
"mean_token_accuracy": 0.1004187822341919,
|
|
"num_tokens": 1689568.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.614206790924072,
|
|
"epoch": 0.8602685231228244,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.3861,
|
|
"mean_token_accuracy": 0.1079413764178753,
|
|
"num_tokens": 1698039.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.502506685256958,
|
|
"epoch": 0.8652411735454998,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.4282,
|
|
"mean_token_accuracy": 0.10273555293679237,
|
|
"num_tokens": 1707659.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.472733020782471,
|
|
"epoch": 0.870213823968175,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.3861,
|
|
"mean_token_accuracy": 0.10378114208579063,
|
|
"num_tokens": 1716952.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.560409307479858,
|
|
"epoch": 0.8751864743908503,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.451,
|
|
"mean_token_accuracy": 0.09781106933951378,
|
|
"num_tokens": 1726323.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.568419170379639,
|
|
"epoch": 0.8801591248135257,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.4508,
|
|
"mean_token_accuracy": 0.09186950996518135,
|
|
"num_tokens": 1736265.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.458628368377686,
|
|
"epoch": 0.8851317752362009,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.3973,
|
|
"mean_token_accuracy": 0.09887660518288613,
|
|
"num_tokens": 1745150.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.540029096603393,
|
|
"epoch": 0.8901044256588762,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5139,
|
|
"mean_token_accuracy": 0.09781100377440452,
|
|
"num_tokens": 1754641.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.446507453918457,
|
|
"epoch": 0.8950770760815515,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.3119,
|
|
"mean_token_accuracy": 0.10973177030682564,
|
|
"num_tokens": 1763468.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.422563886642456,
|
|
"epoch": 0.9000497265042268,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4432,
|
|
"mean_token_accuracy": 0.10403756350278855,
|
|
"num_tokens": 1774845.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.505294227600098,
|
|
"epoch": 0.905022376926902,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.3637,
|
|
"mean_token_accuracy": 0.10615181997418403,
|
|
"num_tokens": 1784606.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.428920269012451,
|
|
"epoch": 0.9099950273495774,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.4024,
|
|
"mean_token_accuracy": 0.10240766182541847,
|
|
"num_tokens": 1793893.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.477800798416138,
|
|
"epoch": 0.9149676777722526,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4009,
|
|
"mean_token_accuracy": 0.1039421871304512,
|
|
"num_tokens": 1804095.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.540898036956787,
|
|
"epoch": 0.9199403281949279,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.3965,
|
|
"mean_token_accuracy": 0.10489938706159592,
|
|
"num_tokens": 1813475.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.317395401000977,
|
|
"epoch": 0.9249129786176031,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.3203,
|
|
"mean_token_accuracy": 0.109952911734581,
|
|
"num_tokens": 1822779.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.5778134822845455,
|
|
"epoch": 0.9298856290402785,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.349,
|
|
"mean_token_accuracy": 0.10500052943825722,
|
|
"num_tokens": 1832239.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.2777961730957035,
|
|
"epoch": 0.9348582794629537,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.3235,
|
|
"mean_token_accuracy": 0.10185129940509796,
|
|
"num_tokens": 1843395.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.460053205490112,
|
|
"epoch": 0.939830929885629,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.442,
|
|
"mean_token_accuracy": 0.10547303482890129,
|
|
"num_tokens": 1853861.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.4669934749603275,
|
|
"epoch": 0.9448035803083044,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.3512,
|
|
"mean_token_accuracy": 0.10421937704086304,
|
|
"num_tokens": 1862992.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.427795505523681,
|
|
"epoch": 0.9497762307309796,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.372,
|
|
"mean_token_accuracy": 0.10107123926281929,
|
|
"num_tokens": 1871684.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.4264805793762205,
|
|
"epoch": 0.9547488811536549,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.3865,
|
|
"mean_token_accuracy": 0.09964759126305581,
|
|
"num_tokens": 1881956.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.41358003616333,
|
|
"epoch": 0.9597215315763302,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.462,
|
|
"mean_token_accuracy": 0.09731132462620735,
|
|
"num_tokens": 1891860.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.419972658157349,
|
|
"epoch": 0.9646941819990055,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.3645,
|
|
"mean_token_accuracy": 0.10411853864789009,
|
|
"num_tokens": 1901521.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.428276777267456,
|
|
"epoch": 0.9696668324216807,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.3269,
|
|
"mean_token_accuracy": 0.10307972654700279,
|
|
"num_tokens": 1910438.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.316250371932983,
|
|
"epoch": 0.9746394828443561,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.3629,
|
|
"mean_token_accuracy": 0.1025395892560482,
|
|
"num_tokens": 1922045.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.4339062690734865,
|
|
"epoch": 0.9796121332670313,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.1867,
|
|
"mean_token_accuracy": 0.11498386934399604,
|
|
"num_tokens": 1930330.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.193427228927613,
|
|
"epoch": 0.9845847836897066,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.1948,
|
|
"mean_token_accuracy": 0.10757651999592781,
|
|
"num_tokens": 1939485.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.486000108718872,
|
|
"epoch": 0.989557434112382,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.2966,
|
|
"mean_token_accuracy": 0.10132434442639351,
|
|
"num_tokens": 1949448.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.363285207748413,
|
|
"epoch": 0.9945300845350572,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.3883,
|
|
"mean_token_accuracy": 0.09904719069600106,
|
|
"num_tokens": 1959923.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9945300845350572,
|
|
"eval_entropy": 7.191312780028664,
|
|
"eval_loss": 7.426286697387695,
|
|
"eval_mean_token_accuracy": 0.09612512418974517,
|
|
"eval_num_tokens": 1959923.0,
|
|
"eval_runtime": 13.1868,
|
|
"eval_samples_per_second": 262.762,
|
|
"eval_steps_per_second": 32.912,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.347610664367676,
|
|
"epoch": 0.9995027349577325,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999997830922735,
|
|
"loss": 7.3185,
|
|
"mean_token_accuracy": 0.10964557453989983,
|
|
"num_tokens": 1969105.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.294963200887044,
|
|
"epoch": 1.0039781203381402,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999989019053515,
|
|
"loss": 7.1344,
|
|
"mean_token_accuracy": 0.10877768033080631,
|
|
"num_tokens": 1979124.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.352720165252686,
|
|
"epoch": 1.0089507707608154,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999973428851536,
|
|
"loss": 7.1593,
|
|
"mean_token_accuracy": 0.1092929445207119,
|
|
"num_tokens": 1987813.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.332399702072143,
|
|
"epoch": 1.0139234211834909,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999951060363766,
|
|
"loss": 7.1376,
|
|
"mean_token_accuracy": 0.10329783409833908,
|
|
"num_tokens": 1996743.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.285995578765869,
|
|
"epoch": 1.0188960716061661,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999921913657592,
|
|
"loss": 7.1417,
|
|
"mean_token_accuracy": 0.11272285059094429,
|
|
"num_tokens": 2006499.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.256630802154541,
|
|
"epoch": 1.0238687220288414,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999885988820821,
|
|
"loss": 7.122,
|
|
"mean_token_accuracy": 0.11043808534741402,
|
|
"num_tokens": 2015103.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.2653522968292235,
|
|
"epoch": 1.0288413724515166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999843285961683,
|
|
"loss": 7.045,
|
|
"mean_token_accuracy": 0.10985710993409156,
|
|
"num_tokens": 2024681.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.284874010086059,
|
|
"epoch": 1.0338140228741919,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999793805208822,
|
|
"loss": 7.1375,
|
|
"mean_token_accuracy": 0.1024433895945549,
|
|
"num_tokens": 2034388.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.21672420501709,
|
|
"epoch": 1.0387866732968671,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999737546711305,
|
|
"loss": 7.065,
|
|
"mean_token_accuracy": 0.1120141789317131,
|
|
"num_tokens": 2042729.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.343318605422974,
|
|
"epoch": 1.0437593237195426,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999674510638618,
|
|
"loss": 7.1751,
|
|
"mean_token_accuracy": 0.10327349603176117,
|
|
"num_tokens": 2051669.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.2701211929321286,
|
|
"epoch": 1.0487319741422179,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999604697180661,
|
|
"loss": 7.1016,
|
|
"mean_token_accuracy": 0.11312837675213813,
|
|
"num_tokens": 2061002.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.16979341506958,
|
|
"epoch": 1.053704624564893,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999528106547759,
|
|
"loss": 7.0115,
|
|
"mean_token_accuracy": 0.11926186233758926,
|
|
"num_tokens": 2070222.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.258184003829956,
|
|
"epoch": 1.0586772749875684,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999444738970644,
|
|
"loss": 7.2081,
|
|
"mean_token_accuracy": 0.10795510858297348,
|
|
"num_tokens": 2079683.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.2904750347137455,
|
|
"epoch": 1.0636499254102436,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999354594700474,
|
|
"loss": 7.1798,
|
|
"mean_token_accuracy": 0.10598527193069458,
|
|
"num_tokens": 2089659.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.248142719268799,
|
|
"epoch": 1.0686225758329189,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999257674008817,
|
|
"loss": 7.0726,
|
|
"mean_token_accuracy": 0.11882116496562958,
|
|
"num_tokens": 2099766.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.186780881881714,
|
|
"epoch": 1.0735952262555943,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999153977187656,
|
|
"loss": 7.2416,
|
|
"mean_token_accuracy": 0.09865990951657296,
|
|
"num_tokens": 2110244.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.36562066078186,
|
|
"epoch": 1.0785678766782696,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000499904350454939,
|
|
"loss": 7.1713,
|
|
"mean_token_accuracy": 0.10669389441609382,
|
|
"num_tokens": 2120068.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.171448564529419,
|
|
"epoch": 1.0835405271009448,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998926256426829,
|
|
"loss": 7.1341,
|
|
"mean_token_accuracy": 0.1101085141301155,
|
|
"num_tokens": 2128862.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.3961996078491214,
|
|
"epoch": 1.08851317752362,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998802233173196,
|
|
"loss": 7.1712,
|
|
"mean_token_accuracy": 0.09955339692533016,
|
|
"num_tokens": 2140618.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.19364185333252,
|
|
"epoch": 1.0934858279462953,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998671435162123,
|
|
"loss": 7.0983,
|
|
"mean_token_accuracy": 0.10813920274376869,
|
|
"num_tokens": 2151890.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.197965812683106,
|
|
"epoch": 1.0984584783689706,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998533862787657,
|
|
"loss": 7.0929,
|
|
"mean_token_accuracy": 0.11835464090108871,
|
|
"num_tokens": 2160862.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.233619642257691,
|
|
"epoch": 1.1034311287916458,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998389516464244,
|
|
"loss": 7.1223,
|
|
"mean_token_accuracy": 0.11691649332642555,
|
|
"num_tokens": 2171631.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.211631155014038,
|
|
"epoch": 1.1084037792143213,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998238396626746,
|
|
"loss": 7.0652,
|
|
"mean_token_accuracy": 0.11038428172469139,
|
|
"num_tokens": 2181564.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.212841987609863,
|
|
"epoch": 1.1133764296369966,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998080503730427,
|
|
"loss": 7.105,
|
|
"mean_token_accuracy": 0.1128804437816143,
|
|
"num_tokens": 2191288.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.188447904586792,
|
|
"epoch": 1.1183490800596718,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997915838250956,
|
|
"loss": 7.0741,
|
|
"mean_token_accuracy": 0.11088593080639839,
|
|
"num_tokens": 2201254.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.137803554534912,
|
|
"epoch": 1.123321730482347,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997744400684406,
|
|
"loss": 7.0288,
|
|
"mean_token_accuracy": 0.11356041058897973,
|
|
"num_tokens": 2210325.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.184323120117187,
|
|
"epoch": 1.1282943809050223,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997566191547251,
|
|
"loss": 7.1396,
|
|
"mean_token_accuracy": 0.10891327261924744,
|
|
"num_tokens": 2220357.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.21702823638916,
|
|
"epoch": 1.1332670313276976,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997381211376364,
|
|
"loss": 7.0674,
|
|
"mean_token_accuracy": 0.12181353569030762,
|
|
"num_tokens": 2229208.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.114451265335083,
|
|
"epoch": 1.138239681750373,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004997189460729019,
|
|
"loss": 7.0615,
|
|
"mean_token_accuracy": 0.11328659504652024,
|
|
"num_tokens": 2238780.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.155827856063842,
|
|
"epoch": 1.1432123321730483,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996990940182884,
|
|
"loss": 7.1093,
|
|
"mean_token_accuracy": 0.1082857072353363,
|
|
"num_tokens": 2249457.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.155684566497802,
|
|
"epoch": 1.1481849825957235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996785650336024,
|
|
"loss": 7.0836,
|
|
"mean_token_accuracy": 0.1123996689915657,
|
|
"num_tokens": 2259565.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.15206995010376,
|
|
"epoch": 1.1531576330183988,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996573591806898,
|
|
"loss": 6.9748,
|
|
"mean_token_accuracy": 0.12142932191491126,
|
|
"num_tokens": 2268301.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.157604646682739,
|
|
"epoch": 1.158130283441074,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996354765234353,
|
|
"loss": 7.1273,
|
|
"mean_token_accuracy": 0.11113866418600082,
|
|
"num_tokens": 2278298.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.121410369873047,
|
|
"epoch": 1.1631029338637493,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996129171277628,
|
|
"loss": 7.0575,
|
|
"mean_token_accuracy": 0.11187752708792686,
|
|
"num_tokens": 2288442.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.113837575912475,
|
|
"epoch": 1.1680755842864246,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995896810616349,
|
|
"loss": 7.0713,
|
|
"mean_token_accuracy": 0.10704104006290435,
|
|
"num_tokens": 2298785.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.166252565383911,
|
|
"epoch": 1.1730482347091,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995657683950528,
|
|
"loss": 7.1323,
|
|
"mean_token_accuracy": 0.10803322568535804,
|
|
"num_tokens": 2308881.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.142189168930054,
|
|
"epoch": 1.1780208851317753,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995411792000559,
|
|
"loss": 7.0221,
|
|
"mean_token_accuracy": 0.11649711951613426,
|
|
"num_tokens": 2318447.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.19552903175354,
|
|
"epoch": 1.1829935355544505,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995159135507218,
|
|
"loss": 7.0222,
|
|
"mean_token_accuracy": 0.12392903566360473,
|
|
"num_tokens": 2327162.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.150506401062012,
|
|
"epoch": 1.1879661859771258,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499489971523166,
|
|
"loss": 7.0646,
|
|
"mean_token_accuracy": 0.10474285408854485,
|
|
"num_tokens": 2337207.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.041934013366699,
|
|
"epoch": 1.192938836399801,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004994633531955415,
|
|
"loss": 7.0686,
|
|
"mean_token_accuracy": 0.10994376018643379,
|
|
"num_tokens": 2347381.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.25481915473938,
|
|
"epoch": 1.1979114868224765,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004994360586480391,
|
|
"loss": 7.107,
|
|
"mean_token_accuracy": 0.11095505580306053,
|
|
"num_tokens": 2357942.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.021949386596679,
|
|
"epoch": 1.2028841372451518,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994080879628864,
|
|
"loss": 7.0248,
|
|
"mean_token_accuracy": 0.11604817807674409,
|
|
"num_tokens": 2367242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.244472694396973,
|
|
"epoch": 1.207856787667827,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004993794412243481,
|
|
"loss": 7.0736,
|
|
"mean_token_accuracy": 0.12035714387893677,
|
|
"num_tokens": 2376344.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.0324465274810795,
|
|
"epoch": 1.2128294380905023,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004993501185187255,
|
|
"loss": 7.0918,
|
|
"mean_token_accuracy": 0.1163498342037201,
|
|
"num_tokens": 2387206.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.19862813949585,
|
|
"epoch": 1.2178020885131775,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993201199343565,
|
|
"loss": 7.122,
|
|
"mean_token_accuracy": 0.11013990938663483,
|
|
"num_tokens": 2397532.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.075332546234131,
|
|
"epoch": 1.2227747389358528,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004992894455616149,
|
|
"loss": 6.9871,
|
|
"mean_token_accuracy": 0.10677824169397354,
|
|
"num_tokens": 2409165.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.139843702316284,
|
|
"epoch": 1.227747389358528,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004992580954929108,
|
|
"loss": 7.0838,
|
|
"mean_token_accuracy": 0.11324674263596535,
|
|
"num_tokens": 2420657.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.090939569473266,
|
|
"epoch": 1.2327200397812033,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004992260698226894,
|
|
"loss": 7.0095,
|
|
"mean_token_accuracy": 0.11596344783902168,
|
|
"num_tokens": 2429936.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.1219744205474855,
|
|
"epoch": 1.2376926902038787,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004991933686474315,
|
|
"loss": 7.0129,
|
|
"mean_token_accuracy": 0.11991648748517036,
|
|
"num_tokens": 2440386.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.074631547927856,
|
|
"epoch": 1.242665340626554,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991599920656529,
|
|
"loss": 7.1722,
|
|
"mean_token_accuracy": 0.10861676037311555,
|
|
"num_tokens": 2449086.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.28495945930481,
|
|
"epoch": 1.2476379910492292,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004991259401779042,
|
|
"loss": 7.1862,
|
|
"mean_token_accuracy": 0.1088343769311905,
|
|
"num_tokens": 2459380.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.109811544418335,
|
|
"epoch": 1.2526106414719045,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990912130867704,
|
|
"loss": 7.0248,
|
|
"mean_token_accuracy": 0.11953925341367722,
|
|
"num_tokens": 2468623.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.142944717407227,
|
|
"epoch": 1.2575832918945797,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004990558108968703,
|
|
"loss": 6.9635,
|
|
"mean_token_accuracy": 0.11530244648456574,
|
|
"num_tokens": 2479817.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.159999370574951,
|
|
"epoch": 1.2625559423172552,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.0004990197337148571,
|
|
"loss": 7.0464,
|
|
"mean_token_accuracy": 0.11831610053777694,
|
|
"num_tokens": 2490544.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.00558180809021,
|
|
"epoch": 1.2675285927399305,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000498982981649417,
|
|
"loss": 6.9865,
|
|
"mean_token_accuracy": 0.1280640684068203,
|
|
"num_tokens": 2500516.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.162667942047119,
|
|
"epoch": 1.2725012431626057,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004989455548112695,
|
|
"loss": 7.1185,
|
|
"mean_token_accuracy": 0.11398841813206673,
|
|
"num_tokens": 2510508.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.103589677810669,
|
|
"epoch": 1.277473893585281,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004989074533131671,
|
|
"loss": 7.0623,
|
|
"mean_token_accuracy": 0.11352467015385628,
|
|
"num_tokens": 2520532.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.144576597213745,
|
|
"epoch": 1.2824465440079562,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004988686772698945,
|
|
"loss": 7.0265,
|
|
"mean_token_accuracy": 0.11438109874725341,
|
|
"num_tokens": 2529912.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.048231172561645,
|
|
"epoch": 1.2874191944306315,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004988292267982689,
|
|
"loss": 7.1069,
|
|
"mean_token_accuracy": 0.10593872368335724,
|
|
"num_tokens": 2540700.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.134967088699341,
|
|
"epoch": 1.2923918448533067,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004987891020171387,
|
|
"loss": 6.9822,
|
|
"mean_token_accuracy": 0.1169443592429161,
|
|
"num_tokens": 2549343.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.020070266723633,
|
|
"epoch": 1.297364495275982,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004987483030473845,
|
|
"loss": 7.0018,
|
|
"mean_token_accuracy": 0.1172388568520546,
|
|
"num_tokens": 2558630.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.051279449462891,
|
|
"epoch": 1.3023371456986574,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004987068300119173,
|
|
"loss": 6.9917,
|
|
"mean_token_accuracy": 0.11961751282215119,
|
|
"num_tokens": 2568559.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.104311180114746,
|
|
"epoch": 1.3073097961213327,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004986646830356792,
|
|
"loss": 7.0531,
|
|
"mean_token_accuracy": 0.11741689294576645,
|
|
"num_tokens": 2578088.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.993081092834473,
|
|
"epoch": 1.312282446544008,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004986218622456423,
|
|
"loss": 7.0065,
|
|
"mean_token_accuracy": 0.12129950895905495,
|
|
"num_tokens": 2588518.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.208339786529541,
|
|
"epoch": 1.3172550969666832,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000498578367770809,
|
|
"loss": 7.1257,
|
|
"mean_token_accuracy": 0.10805041193962098,
|
|
"num_tokens": 2598850.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.123275184631348,
|
|
"epoch": 1.3222277473893587,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004985341997422111,
|
|
"loss": 7.1147,
|
|
"mean_token_accuracy": 0.10578964054584503,
|
|
"num_tokens": 2610667.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.075258445739746,
|
|
"epoch": 1.327200397812034,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004984893582929093,
|
|
"loss": 6.9709,
|
|
"mean_token_accuracy": 0.12416771575808525,
|
|
"num_tokens": 2620430.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.975117254257202,
|
|
"epoch": 1.3321730482347092,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004984438435579933,
|
|
"loss": 6.8335,
|
|
"mean_token_accuracy": 0.12780645191669465,
|
|
"num_tokens": 2629433.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.016169691085816,
|
|
"epoch": 1.3371456986573844,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004983976556745812,
|
|
"loss": 7.0215,
|
|
"mean_token_accuracy": 0.10864698886871338,
|
|
"num_tokens": 2639611.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.162472248077393,
|
|
"epoch": 1.3421183490800597,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000498350794781819,
|
|
"loss": 7.0179,
|
|
"mean_token_accuracy": 0.11265038326382637,
|
|
"num_tokens": 2649494.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.012684392929077,
|
|
"epoch": 1.347090999502735,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004983032610208801,
|
|
"loss": 6.9364,
|
|
"mean_token_accuracy": 0.11880568042397499,
|
|
"num_tokens": 2658723.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.9948114395141605,
|
|
"epoch": 1.3520636499254102,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004982550545349649,
|
|
"loss": 6.9928,
|
|
"mean_token_accuracy": 0.11781855300068855,
|
|
"num_tokens": 2668530.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.124820041656494,
|
|
"epoch": 1.3570363003480854,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004982061754693008,
|
|
"loss": 7.0607,
|
|
"mean_token_accuracy": 0.11784527972340583,
|
|
"num_tokens": 2678788.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.024953413009643,
|
|
"epoch": 1.362008950770761,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004981566239711412,
|
|
"loss": 6.9529,
|
|
"mean_token_accuracy": 0.11857759058475495,
|
|
"num_tokens": 2688308.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.848841857910156,
|
|
"epoch": 1.3669816011934361,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 0.0004981064001897651,
|
|
"loss": 6.8951,
|
|
"mean_token_accuracy": 0.12636217325925828,
|
|
"num_tokens": 2698146.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.089616441726685,
|
|
"epoch": 1.3719542516161114,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004980555042764772,
|
|
"loss": 6.9735,
|
|
"mean_token_accuracy": 0.11414305567741394,
|
|
"num_tokens": 2707336.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.029972267150879,
|
|
"epoch": 1.3769269020387866,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004980039363846069,
|
|
"loss": 6.9462,
|
|
"mean_token_accuracy": 0.12458088025450706,
|
|
"num_tokens": 2717573.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.040216732025146,
|
|
"epoch": 1.381899552461462,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497951696669508,
|
|
"loss": 6.9358,
|
|
"mean_token_accuracy": 0.12328411564230919,
|
|
"num_tokens": 2727488.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.927717638015747,
|
|
"epoch": 1.3868722028841374,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004978987852885579,
|
|
"loss": 6.9371,
|
|
"mean_token_accuracy": 0.12751979753375053,
|
|
"num_tokens": 2738010.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.080646276473999,
|
|
"epoch": 1.3918448533068126,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004978452024011582,
|
|
"loss": 6.9998,
|
|
"mean_token_accuracy": 0.11901936903595925,
|
|
"num_tokens": 2746984.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.0536294937133786,
|
|
"epoch": 1.3968175037294879,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497790948168733,
|
|
"loss": 6.9911,
|
|
"mean_token_accuracy": 0.1136419303715229,
|
|
"num_tokens": 2756578.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.9825427532196045,
|
|
"epoch": 1.4017901541521631,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004977360227547288,
|
|
"loss": 6.9295,
|
|
"mean_token_accuracy": 0.12150286510586739,
|
|
"num_tokens": 2765773.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.9939939975738525,
|
|
"epoch": 1.4067628045748384,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004976804263246146,
|
|
"loss": 6.931,
|
|
"mean_token_accuracy": 0.12239946499466896,
|
|
"num_tokens": 2775002.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.114528703689575,
|
|
"epoch": 1.4117354549975136,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004976241590458803,
|
|
"loss": 6.995,
|
|
"mean_token_accuracy": 0.1176114745438099,
|
|
"num_tokens": 2784682.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.918878602981567,
|
|
"epoch": 1.4167081054201889,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004975672210880372,
|
|
"loss": 6.9092,
|
|
"mean_token_accuracy": 0.12430581822991371,
|
|
"num_tokens": 2794371.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.0797919750213625,
|
|
"epoch": 1.4216807558428641,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004975096126226171,
|
|
"loss": 6.9706,
|
|
"mean_token_accuracy": 0.11535287648439407,
|
|
"num_tokens": 2803840.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.925989580154419,
|
|
"epoch": 1.4266534062655396,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974513338231715,
|
|
"loss": 6.881,
|
|
"mean_token_accuracy": 0.12455180361866951,
|
|
"num_tokens": 2813536.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.95003662109375,
|
|
"epoch": 1.4316260566882149,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004973923848652715,
|
|
"loss": 6.9413,
|
|
"mean_token_accuracy": 0.12003065943717957,
|
|
"num_tokens": 2822741.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.054536437988281,
|
|
"epoch": 1.43659870711089,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004973327659265073,
|
|
"loss": 6.91,
|
|
"mean_token_accuracy": 0.1195470042526722,
|
|
"num_tokens": 2832208.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.83617639541626,
|
|
"epoch": 1.4415713575335654,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004972724771864874,
|
|
"loss": 6.8016,
|
|
"mean_token_accuracy": 0.1272654265165329,
|
|
"num_tokens": 2841672.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.944074249267578,
|
|
"epoch": 1.4465440079562406,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004972115188268378,
|
|
"loss": 6.9011,
|
|
"mean_token_accuracy": 0.12446089461445808,
|
|
"num_tokens": 2852323.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.0025616645812985,
|
|
"epoch": 1.451516658378916,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004971498910312024,
|
|
"loss": 6.9636,
|
|
"mean_token_accuracy": 0.12246960997581482,
|
|
"num_tokens": 2861863.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.032646703720093,
|
|
"epoch": 1.4564893088015913,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004970875939852412,
|
|
"loss": 6.8999,
|
|
"mean_token_accuracy": 0.12254282459616661,
|
|
"num_tokens": 2871608.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.945109891891479,
|
|
"epoch": 1.4614619592242666,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497024627876631,
|
|
"loss": 6.8532,
|
|
"mean_token_accuracy": 0.12762577533721925,
|
|
"num_tokens": 2882091.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.951807451248169,
|
|
"epoch": 1.4664346096469418,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004969609928950639,
|
|
"loss": 6.8377,
|
|
"mean_token_accuracy": 0.13220236003398894,
|
|
"num_tokens": 2892499.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.008027744293213,
|
|
"epoch": 1.471407260069617,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004968966892322468,
|
|
"loss": 6.8817,
|
|
"mean_token_accuracy": 0.11723986715078354,
|
|
"num_tokens": 2902208.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.939340019226075,
|
|
"epoch": 1.4763799104922923,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004968317170819018,
|
|
"loss": 6.8829,
|
|
"mean_token_accuracy": 0.12179240882396698,
|
|
"num_tokens": 2911196.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.918968582153321,
|
|
"epoch": 1.4813525609149676,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004967660766397642,
|
|
"loss": 6.9252,
|
|
"mean_token_accuracy": 0.11802939847111701,
|
|
"num_tokens": 2921339.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.000785255432129,
|
|
"epoch": 1.4863252113376428,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004966997681035829,
|
|
"loss": 6.8658,
|
|
"mean_token_accuracy": 0.12102236896753311,
|
|
"num_tokens": 2930795.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.9171936988830565,
|
|
"epoch": 1.4912978617603183,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004966327916731195,
|
|
"loss": 6.8298,
|
|
"mean_token_accuracy": 0.12037147507071495,
|
|
"num_tokens": 2939647.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.4912978617603183,
|
|
"eval_entropy": 6.852062745028377,
|
|
"eval_loss": 7.132150650024414,
|
|
"eval_mean_token_accuracy": 0.11200833488427411,
|
|
"eval_num_tokens": 2939647.0,
|
|
"eval_runtime": 13.99,
|
|
"eval_samples_per_second": 247.676,
|
|
"eval_steps_per_second": 31.022,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.940325403213501,
|
|
"epoch": 1.4962705121829936,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004965651475501476,
|
|
"loss": 6.8749,
|
|
"mean_token_accuracy": 0.1193210557103157,
|
|
"num_tokens": 2948365.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.860566091537476,
|
|
"epoch": 1.5012431626056688,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004964968359384524,
|
|
"loss": 6.8852,
|
|
"mean_token_accuracy": 0.11689071506261825,
|
|
"num_tokens": 2957447.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.01457405090332,
|
|
"epoch": 1.506215813028344,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964278570438299,
|
|
"loss": 6.8816,
|
|
"mean_token_accuracy": 0.12147188484668732,
|
|
"num_tokens": 2965938.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.96983060836792,
|
|
"epoch": 1.5111884634510195,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004963582110740865,
|
|
"loss": 6.8754,
|
|
"mean_token_accuracy": 0.12140716314315796,
|
|
"num_tokens": 2975830.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.919324254989624,
|
|
"epoch": 1.5161611138736948,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004962878982390381,
|
|
"loss": 6.8763,
|
|
"mean_token_accuracy": 0.1240736149251461,
|
|
"num_tokens": 2985391.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.987160968780517,
|
|
"epoch": 1.52113376429637,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004962169187505097,
|
|
"loss": 6.8631,
|
|
"mean_token_accuracy": 0.13275940716266632,
|
|
"num_tokens": 2994397.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.920234823226929,
|
|
"epoch": 1.5261064147190453,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004961452728223343,
|
|
"loss": 6.8986,
|
|
"mean_token_accuracy": 0.11727899610996247,
|
|
"num_tokens": 3005085.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.903799724578858,
|
|
"epoch": 1.5310790651417205,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004960729606703535,
|
|
"loss": 6.8895,
|
|
"mean_token_accuracy": 0.12234467566013336,
|
|
"num_tokens": 3015500.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.036432695388794,
|
|
"epoch": 1.5360517155643958,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004959999825124147,
|
|
"loss": 6.9961,
|
|
"mean_token_accuracy": 0.11839420944452286,
|
|
"num_tokens": 3024862.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.996464967727661,
|
|
"epoch": 1.541024365987071,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000495926338568373,
|
|
"loss": 6.8478,
|
|
"mean_token_accuracy": 0.1252470299601555,
|
|
"num_tokens": 3034775.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.878435945510864,
|
|
"epoch": 1.5459970164097463,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004958520290600884,
|
|
"loss": 6.9032,
|
|
"mean_token_accuracy": 0.11878609433770179,
|
|
"num_tokens": 3044048.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.034942626953125,
|
|
"epoch": 1.5509696668324215,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004957770542114262,
|
|
"loss": 6.9101,
|
|
"mean_token_accuracy": 0.12281852588057518,
|
|
"num_tokens": 3054237.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.741971015930176,
|
|
"epoch": 1.555942317255097,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004957014142482563,
|
|
"loss": 6.7204,
|
|
"mean_token_accuracy": 0.13250551000237465,
|
|
"num_tokens": 3064787.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.8905096530914305,
|
|
"epoch": 1.5609149676777723,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004956251093984521,
|
|
"loss": 6.8872,
|
|
"mean_token_accuracy": 0.12106038108468056,
|
|
"num_tokens": 3074002.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.75254716873169,
|
|
"epoch": 1.5658876181004475,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004955481398918902,
|
|
"loss": 6.7553,
|
|
"mean_token_accuracy": 0.12440378069877625,
|
|
"num_tokens": 3084270.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.972566223144531,
|
|
"epoch": 1.5708602685231228,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004954705059604495,
|
|
"loss": 6.8731,
|
|
"mean_token_accuracy": 0.12709827348589897,
|
|
"num_tokens": 3095236.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.923832988739013,
|
|
"epoch": 1.5758329189457982,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004953922078380104,
|
|
"loss": 6.8607,
|
|
"mean_token_accuracy": 0.12445903047919274,
|
|
"num_tokens": 3104912.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.8252942085266115,
|
|
"epoch": 1.5808055693684735,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004953132457604544,
|
|
"loss": 6.8662,
|
|
"mean_token_accuracy": 0.1228304535150528,
|
|
"num_tokens": 3114782.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.066912460327148,
|
|
"epoch": 1.5857782197911487,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004952336199656632,
|
|
"loss": 6.8962,
|
|
"mean_token_accuracy": 0.12412459626793862,
|
|
"num_tokens": 3123606.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.792387866973877,
|
|
"epoch": 1.590750870213824,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004951533306935181,
|
|
"loss": 6.7004,
|
|
"mean_token_accuracy": 0.13109694942831993,
|
|
"num_tokens": 3133343.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.974667024612427,
|
|
"epoch": 1.5957235206364992,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004950723781858991,
|
|
"loss": 6.9609,
|
|
"mean_token_accuracy": 0.11378122866153717,
|
|
"num_tokens": 3143137.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.949428081512451,
|
|
"epoch": 1.6006961710591745,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004949907626866844,
|
|
"loss": 6.8217,
|
|
"mean_token_accuracy": 0.12410593032836914,
|
|
"num_tokens": 3152595.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.86395034790039,
|
|
"epoch": 1.6056688214818498,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004949084844417493,
|
|
"loss": 6.8927,
|
|
"mean_token_accuracy": 0.12164935991168022,
|
|
"num_tokens": 3162333.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.856705522537231,
|
|
"epoch": 1.610641471904525,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004948255436989657,
|
|
"loss": 6.7865,
|
|
"mean_token_accuracy": 0.1268024265766144,
|
|
"num_tokens": 3172765.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.957909870147705,
|
|
"epoch": 1.6156141223272003,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004947419407082017,
|
|
"loss": 6.9398,
|
|
"mean_token_accuracy": 0.11589413285255432,
|
|
"num_tokens": 3183613.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.00533013343811,
|
|
"epoch": 1.6205867727498757,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004946576757213202,
|
|
"loss": 6.8487,
|
|
"mean_token_accuracy": 0.12260088175535203,
|
|
"num_tokens": 3193020.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.890829181671142,
|
|
"epoch": 1.625559423172551,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004945727489921784,
|
|
"loss": 6.826,
|
|
"mean_token_accuracy": 0.12254872098565102,
|
|
"num_tokens": 3202845.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.820006084442139,
|
|
"epoch": 1.6305320735952262,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004944871607766273,
|
|
"loss": 6.7792,
|
|
"mean_token_accuracy": 0.12128874883055688,
|
|
"num_tokens": 3211195.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.932735776901245,
|
|
"epoch": 1.6355047240179017,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004944009113325104,
|
|
"loss": 6.8654,
|
|
"mean_token_accuracy": 0.12427843883633613,
|
|
"num_tokens": 3221549.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.858042240142822,
|
|
"epoch": 1.640477374440577,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004943140009196635,
|
|
"loss": 6.729,
|
|
"mean_token_accuracy": 0.13378771319985389,
|
|
"num_tokens": 3232206.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.925443458557129,
|
|
"epoch": 1.6454500248632522,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004942264297999135,
|
|
"loss": 6.8271,
|
|
"mean_token_accuracy": 0.11983336359262467,
|
|
"num_tokens": 3241613.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.745822238922119,
|
|
"epoch": 1.6504226752859275,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004941381982370778,
|
|
"loss": 6.7891,
|
|
"mean_token_accuracy": 0.12697399258613587,
|
|
"num_tokens": 3251968.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.8707013607025145,
|
|
"epoch": 1.6553953257086027,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004940493064969633,
|
|
"loss": 6.7789,
|
|
"mean_token_accuracy": 0.12398573905229568,
|
|
"num_tokens": 3261969.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.882634401321411,
|
|
"epoch": 1.660367976131278,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000493959754847366,
|
|
"loss": 6.8213,
|
|
"mean_token_accuracy": 0.12630714252591133,
|
|
"num_tokens": 3270894.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.847609519958496,
|
|
"epoch": 1.6653406265539532,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004938695435580698,
|
|
"loss": 6.872,
|
|
"mean_token_accuracy": 0.11402463540434837,
|
|
"num_tokens": 3280949.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.005016040802002,
|
|
"epoch": 1.6703132769766285,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004937786729008459,
|
|
"loss": 6.8324,
|
|
"mean_token_accuracy": 0.1208095520734787,
|
|
"num_tokens": 3291449.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.787694501876831,
|
|
"epoch": 1.6752859273993037,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004936871431494517,
|
|
"loss": 6.8336,
|
|
"mean_token_accuracy": 0.12228364273905754,
|
|
"num_tokens": 3301632.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.87408037185669,
|
|
"epoch": 1.680258577821979,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004935949545796307,
|
|
"loss": 6.6941,
|
|
"mean_token_accuracy": 0.13019528537988662,
|
|
"num_tokens": 3310802.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.789710855484008,
|
|
"epoch": 1.6852312282446544,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004935021074691106,
|
|
"loss": 6.7616,
|
|
"mean_token_accuracy": 0.12533998340368271,
|
|
"num_tokens": 3320833.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.853605461120606,
|
|
"epoch": 1.6902038786673297,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004934086020976032,
|
|
"loss": 6.7535,
|
|
"mean_token_accuracy": 0.13115666210651397,
|
|
"num_tokens": 3330030.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.6503077983856205,
|
|
"epoch": 1.695176529090005,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004933144387468037,
|
|
"loss": 6.5465,
|
|
"mean_token_accuracy": 0.14418469816446305,
|
|
"num_tokens": 3338745.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.830546045303345,
|
|
"epoch": 1.7001491795126804,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004932196177003891,
|
|
"loss": 6.7822,
|
|
"mean_token_accuracy": 0.12927022501826285,
|
|
"num_tokens": 3348067.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.741507577896118,
|
|
"epoch": 1.7051218299353557,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000493124139244018,
|
|
"loss": 6.6715,
|
|
"mean_token_accuracy": 0.13641394078731536,
|
|
"num_tokens": 3357061.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.826949787139893,
|
|
"epoch": 1.710094480358031,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004930280036653294,
|
|
"loss": 6.7084,
|
|
"mean_token_accuracy": 0.12743898034095763,
|
|
"num_tokens": 3366573.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.825648784637451,
|
|
"epoch": 1.7150671307807062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492931211253942,
|
|
"loss": 6.7869,
|
|
"mean_token_accuracy": 0.12010818794369697,
|
|
"num_tokens": 3376996.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.921758508682251,
|
|
"epoch": 1.7200397812033814,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004928337623014535,
|
|
"loss": 6.8066,
|
|
"mean_token_accuracy": 0.11920888870954513,
|
|
"num_tokens": 3387615.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.708018970489502,
|
|
"epoch": 1.7250124316260567,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000492735657101439,
|
|
"loss": 6.5862,
|
|
"mean_token_accuracy": 0.14756272062659265,
|
|
"num_tokens": 3397578.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.8108823776245115,
|
|
"epoch": 1.729985082048732,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004926368959494513,
|
|
"loss": 6.7752,
|
|
"mean_token_accuracy": 0.12232266888022422,
|
|
"num_tokens": 3406697.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.899456453323364,
|
|
"epoch": 1.7349577324714072,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004925374791430186,
|
|
"loss": 6.7434,
|
|
"mean_token_accuracy": 0.13430135771632196,
|
|
"num_tokens": 3416631.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.785489845275879,
|
|
"epoch": 1.7399303828940824,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004924374069816448,
|
|
"loss": 6.8342,
|
|
"mean_token_accuracy": 0.11735361218452453,
|
|
"num_tokens": 3426564.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.921190643310547,
|
|
"epoch": 1.7449030333167577,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004923366797668081,
|
|
"loss": 6.8518,
|
|
"mean_token_accuracy": 0.12866259068250657,
|
|
"num_tokens": 3436200.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.818837022781372,
|
|
"epoch": 1.7498756837394331,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004922352978019597,
|
|
"loss": 6.8227,
|
|
"mean_token_accuracy": 0.12548279836773873,
|
|
"num_tokens": 3446460.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.8263813018798825,
|
|
"epoch": 1.7548483341621084,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0004921332613925239,
|
|
"loss": 6.7014,
|
|
"mean_token_accuracy": 0.13551281541585922,
|
|
"num_tokens": 3457629.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.889324426651001,
|
|
"epoch": 1.7598209845847836,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004920305708458964,
|
|
"loss": 6.8588,
|
|
"mean_token_accuracy": 0.12121690437197685,
|
|
"num_tokens": 3467754.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.784916973114013,
|
|
"epoch": 1.7647936350074591,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004919272264714431,
|
|
"loss": 6.7962,
|
|
"mean_token_accuracy": 0.1267058253288269,
|
|
"num_tokens": 3477094.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.851461887359619,
|
|
"epoch": 1.7697662854301344,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004918232285805004,
|
|
"loss": 6.7484,
|
|
"mean_token_accuracy": 0.12927899286150932,
|
|
"num_tokens": 3487276.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.81771240234375,
|
|
"epoch": 1.7747389358528096,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004917185774863727,
|
|
"loss": 6.7621,
|
|
"mean_token_accuracy": 0.1282905027270317,
|
|
"num_tokens": 3498118.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.6943212985992435,
|
|
"epoch": 1.7797115862754849,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004916132735043329,
|
|
"loss": 6.572,
|
|
"mean_token_accuracy": 0.13395634964108466,
|
|
"num_tokens": 3508286.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.744240140914917,
|
|
"epoch": 1.7846842366981601,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004915073169516204,
|
|
"loss": 6.7061,
|
|
"mean_token_accuracy": 0.12887117713689805,
|
|
"num_tokens": 3517979.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.766994142532349,
|
|
"epoch": 1.7896568871208354,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004914007081474406,
|
|
"loss": 6.735,
|
|
"mean_token_accuracy": 0.1293118454515934,
|
|
"num_tokens": 3528003.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.927664470672608,
|
|
"epoch": 1.7946295375435106,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004912934474129641,
|
|
"loss": 6.7964,
|
|
"mean_token_accuracy": 0.12800249010324477,
|
|
"num_tokens": 3537932.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.774818801879883,
|
|
"epoch": 1.7996021879661859,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004911855350713251,
|
|
"loss": 6.9175,
|
|
"mean_token_accuracy": 0.12320146858692169,
|
|
"num_tokens": 3548584.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.923383092880249,
|
|
"epoch": 1.8045748383888611,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004910769714476214,
|
|
"loss": 6.7202,
|
|
"mean_token_accuracy": 0.13341662883758545,
|
|
"num_tokens": 3558117.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.788570928573608,
|
|
"epoch": 1.8095474888115366,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004909677568689122,
|
|
"loss": 6.8639,
|
|
"mean_token_accuracy": 0.11817813217639923,
|
|
"num_tokens": 3569226.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.881636714935302,
|
|
"epoch": 1.8145201392342118,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004908578916642183,
|
|
"loss": 6.7298,
|
|
"mean_token_accuracy": 0.12382532209157944,
|
|
"num_tokens": 3578375.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.797626495361328,
|
|
"epoch": 1.819492789656887,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004907473761645205,
|
|
"loss": 6.7322,
|
|
"mean_token_accuracy": 0.12980048581957818,
|
|
"num_tokens": 3587356.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.860454750061035,
|
|
"epoch": 1.8244654400795626,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004906362107027583,
|
|
"loss": 6.7971,
|
|
"mean_token_accuracy": 0.12274517342448235,
|
|
"num_tokens": 3596785.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.801642417907715,
|
|
"epoch": 1.8294380905022378,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004905243956138298,
|
|
"loss": 6.6906,
|
|
"mean_token_accuracy": 0.13151210248470308,
|
|
"num_tokens": 3605357.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.791120433807373,
|
|
"epoch": 1.834410740924913,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004904119312345896,
|
|
"loss": 6.7199,
|
|
"mean_token_accuracy": 0.13134362250566484,
|
|
"num_tokens": 3614979.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.829075860977173,
|
|
"epoch": 1.8393833913475883,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004902988179038492,
|
|
"loss": 6.6986,
|
|
"mean_token_accuracy": 0.1295736663043499,
|
|
"num_tokens": 3624594.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.755134248733521,
|
|
"epoch": 1.8443560417702636,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000490185055962374,
|
|
"loss": 6.7103,
|
|
"mean_token_accuracy": 0.12882627472281455,
|
|
"num_tokens": 3633998.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.782251310348511,
|
|
"epoch": 1.8493286921929388,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004900706457528845,
|
|
"loss": 6.6663,
|
|
"mean_token_accuracy": 0.13423041701316835,
|
|
"num_tokens": 3642406.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.716598987579346,
|
|
"epoch": 1.854301342615614,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004899555876200535,
|
|
"loss": 6.7352,
|
|
"mean_token_accuracy": 0.1293344371020794,
|
|
"num_tokens": 3652905.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.799939155578613,
|
|
"epoch": 1.8592739930382893,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004898398819105059,
|
|
"loss": 6.6589,
|
|
"mean_token_accuracy": 0.1286543920636177,
|
|
"num_tokens": 3662518.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.814232110977173,
|
|
"epoch": 1.8642466434609646,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004897235289728178,
|
|
"loss": 6.7133,
|
|
"mean_token_accuracy": 0.13452256545424462,
|
|
"num_tokens": 3672102.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.819967555999756,
|
|
"epoch": 1.8692192938836398,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004896065291575145,
|
|
"loss": 6.7492,
|
|
"mean_token_accuracy": 0.12764293029904367,
|
|
"num_tokens": 3681821.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.93118987083435,
|
|
"epoch": 1.8741919443063153,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004894888828170708,
|
|
"loss": 6.841,
|
|
"mean_token_accuracy": 0.1225848525762558,
|
|
"num_tokens": 3690753.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.748576593399048,
|
|
"epoch": 1.8791645947289906,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004893705903059087,
|
|
"loss": 6.7866,
|
|
"mean_token_accuracy": 0.1224599875509739,
|
|
"num_tokens": 3701694.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.813471746444702,
|
|
"epoch": 1.8841372451516658,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004892516519803971,
|
|
"loss": 6.6983,
|
|
"mean_token_accuracy": 0.12926713824272157,
|
|
"num_tokens": 3710894.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.690635108947754,
|
|
"epoch": 1.8891098955743413,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004891320681988504,
|
|
"loss": 6.6549,
|
|
"mean_token_accuracy": 0.1363897442817688,
|
|
"num_tokens": 3720081.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.800876712799072,
|
|
"epoch": 1.8940825459970165,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004890118393215277,
|
|
"loss": 6.7613,
|
|
"mean_token_accuracy": 0.1292950838804245,
|
|
"num_tokens": 3731168.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.758112621307373,
|
|
"epoch": 1.8990551964196918,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004888909657106311,
|
|
"loss": 6.5677,
|
|
"mean_token_accuracy": 0.1415342666208744,
|
|
"num_tokens": 3740533.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.704781341552734,
|
|
"epoch": 1.904027846842367,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004887694477303055,
|
|
"loss": 6.7273,
|
|
"mean_token_accuracy": 0.12590624764561653,
|
|
"num_tokens": 3750523.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.819173574447632,
|
|
"epoch": 1.9090004972650423,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004886472857466368,
|
|
"loss": 6.7596,
|
|
"mean_token_accuracy": 0.12585822269320487,
|
|
"num_tokens": 3760323.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.618635749816894,
|
|
"epoch": 1.9139731476877175,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004885244801276509,
|
|
"loss": 6.5051,
|
|
"mean_token_accuracy": 0.14932959750294686,
|
|
"num_tokens": 3770970.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.7610045909881595,
|
|
"epoch": 1.9189457981103928,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004884010312433131,
|
|
"loss": 6.7436,
|
|
"mean_token_accuracy": 0.12835264354944229,
|
|
"num_tokens": 3780810.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.767989683151245,
|
|
"epoch": 1.923918448533068,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048827693946552615,
|
|
"loss": 6.6576,
|
|
"mean_token_accuracy": 0.13518533408641814,
|
|
"num_tokens": 3791461.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.646088027954102,
|
|
"epoch": 1.9288910989557433,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048815220516813004,
|
|
"loss": 6.4413,
|
|
"mean_token_accuracy": 0.14974167719483375,
|
|
"num_tokens": 3800353.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.701562738418579,
|
|
"epoch": 1.9338637493784185,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004880268287269,
|
|
"loss": 6.6449,
|
|
"mean_token_accuracy": 0.13712267428636551,
|
|
"num_tokens": 3810633.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.8101723194122314,
|
|
"epoch": 1.938836399801094,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048790081051954605,
|
|
"loss": 6.7966,
|
|
"mean_token_accuracy": 0.12751081138849257,
|
|
"num_tokens": 3820254.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.812269592285157,
|
|
"epoch": 1.9438090502237693,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004877741509257116,
|
|
"loss": 6.7661,
|
|
"mean_token_accuracy": 0.12658249586820602,
|
|
"num_tokens": 3829976.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.701629734039306,
|
|
"epoch": 1.9487817006464445,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004876468503269721,
|
|
"loss": 6.6979,
|
|
"mean_token_accuracy": 0.12903292253613471,
|
|
"num_tokens": 3840593.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.77159366607666,
|
|
"epoch": 1.95375435106912,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048751890910683434,
|
|
"loss": 6.6865,
|
|
"mean_token_accuracy": 0.13638681024312974,
|
|
"num_tokens": 3850063.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.8100159645080565,
|
|
"epoch": 1.9587270014917952,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048739032765073487,
|
|
"loss": 6.7331,
|
|
"mean_token_accuracy": 0.13090887293219566,
|
|
"num_tokens": 3860132.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.7325685024261475,
|
|
"epoch": 1.9636996519144705,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048726110634603913,
|
|
"loss": 6.6763,
|
|
"mean_token_accuracy": 0.13326457887887955,
|
|
"num_tokens": 3869914.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.655536556243897,
|
|
"epoch": 1.9686723023371457,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048713124558204013,
|
|
"loss": 6.6319,
|
|
"mean_token_accuracy": 0.14084189608693123,
|
|
"num_tokens": 3879825.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.8169543743133545,
|
|
"epoch": 1.973644952759821,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004870007457499573,
|
|
"loss": 6.7123,
|
|
"mean_token_accuracy": 0.13578568771481514,
|
|
"num_tokens": 3889318.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.753716945648193,
|
|
"epoch": 1.9786176031824962,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048686960724293537,
|
|
"loss": 6.69,
|
|
"mean_token_accuracy": 0.12853146120905876,
|
|
"num_tokens": 3899017.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.7552789688110355,
|
|
"epoch": 1.9835902536051715,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004867378304560432,
|
|
"loss": 6.6439,
|
|
"mean_token_accuracy": 0.13850356340408326,
|
|
"num_tokens": 3907766.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.752278661727905,
|
|
"epoch": 1.9885629040278467,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004866054157862723,
|
|
"loss": 6.6308,
|
|
"mean_token_accuracy": 0.13319557160139084,
|
|
"num_tokens": 3916323.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9885629040278467,
|
|
"eval_entropy": 6.545219276357906,
|
|
"eval_loss": 6.889737606048584,
|
|
"eval_mean_token_accuracy": 0.12097798801168869,
|
|
"eval_num_tokens": 3916323.0,
|
|
"eval_runtime": 8.4746,
|
|
"eval_samples_per_second": 408.867,
|
|
"eval_steps_per_second": 51.212,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.678959369659424,
|
|
"epoch": 1.993535554450522,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004864723636325361,
|
|
"loss": 6.7119,
|
|
"mean_token_accuracy": 0.13230316191911698,
|
|
"num_tokens": 3927639.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.772380590438843,
|
|
"epoch": 1.9985082048731975,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00048633867439566855,
|
|
"loss": 6.6324,
|
|
"mean_token_accuracy": 0.12921209931373595,
|
|
"num_tokens": 3937452.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.814700921376546,
|
|
"epoch": 2.002983590253605,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004862043484784227,
|
|
"loss": 6.5724,
|
|
"mean_token_accuracy": 0.1322319499320454,
|
|
"num_tokens": 3946737.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.715101957321167,
|
|
"epoch": 2.0079562406762803,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048606938628546975,
|
|
"loss": 6.363,
|
|
"mean_token_accuracy": 0.13461797833442687,
|
|
"num_tokens": 3957541.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.728018283843994,
|
|
"epoch": 2.0129288910989556,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004859337882233978,
|
|
"loss": 6.428,
|
|
"mean_token_accuracy": 0.13036251217126846,
|
|
"num_tokens": 3967621.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.734221696853638,
|
|
"epoch": 2.017901541521631,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004857975547007106,
|
|
"loss": 6.4336,
|
|
"mean_token_accuracy": 0.1381743036210537,
|
|
"num_tokens": 3977920.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.498066568374634,
|
|
"epoch": 2.0228741919443065,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048566068612782595,
|
|
"loss": 6.0909,
|
|
"mean_token_accuracy": 0.16107537150382994,
|
|
"num_tokens": 3988128.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.663156366348266,
|
|
"epoch": 2.0278468423669818,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048552318291707533,
|
|
"loss": 6.3055,
|
|
"mean_token_accuracy": 0.1394015684723854,
|
|
"num_tokens": 3998467.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.712711954116822,
|
|
"epoch": 2.032819492789657,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048538504548270177,
|
|
"loss": 6.2691,
|
|
"mean_token_accuracy": 0.13480257615447044,
|
|
"num_tokens": 4007513.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.4561097621917725,
|
|
"epoch": 2.0377921432123323,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048524627424085913,
|
|
"loss": 6.3182,
|
|
"mean_token_accuracy": 0.14225988388061522,
|
|
"num_tokens": 4016538.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.620581436157226,
|
|
"epoch": 2.0427647936350075,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048510686960961057,
|
|
"loss": 6.2479,
|
|
"mean_token_accuracy": 0.14379681423306465,
|
|
"num_tokens": 4026850.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.647997951507568,
|
|
"epoch": 2.0477374440576828,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004849668320089276,
|
|
"loss": 6.2441,
|
|
"mean_token_accuracy": 0.14394547790288925,
|
|
"num_tokens": 4037105.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.51926817893982,
|
|
"epoch": 2.052710094480358,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048482616186068826,
|
|
"loss": 6.2907,
|
|
"mean_token_accuracy": 0.14822532311081887,
|
|
"num_tokens": 4045694.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.525151157379151,
|
|
"epoch": 2.0576827449030333,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048468485958867664,
|
|
"loss": 6.1816,
|
|
"mean_token_accuracy": 0.15257717669010162,
|
|
"num_tokens": 4056425.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.557491683959961,
|
|
"epoch": 2.0626553953257085,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048454292561858083,
|
|
"loss": 6.2184,
|
|
"mean_token_accuracy": 0.1503872439265251,
|
|
"num_tokens": 4066257.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.5382547855377195,
|
|
"epoch": 2.0676280457483838,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004844003603779921,
|
|
"loss": 6.2987,
|
|
"mean_token_accuracy": 0.1429552063345909,
|
|
"num_tokens": 4076835.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.643266391754151,
|
|
"epoch": 2.072600696171059,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004842571642964036,
|
|
"loss": 6.4089,
|
|
"mean_token_accuracy": 0.1361112542450428,
|
|
"num_tokens": 4087644.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.6248595237731935,
|
|
"epoch": 2.0775733465937343,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048411333780520876,
|
|
"loss": 6.3311,
|
|
"mean_token_accuracy": 0.14037291407585145,
|
|
"num_tokens": 4097040.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.586101293563843,
|
|
"epoch": 2.0825459970164095,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048396888133770034,
|
|
"loss": 6.3254,
|
|
"mean_token_accuracy": 0.13902762457728385,
|
|
"num_tokens": 4106698.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.615466833114624,
|
|
"epoch": 2.087518647439085,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048382379532906897,
|
|
"loss": 6.3175,
|
|
"mean_token_accuracy": 0.1463276758790016,
|
|
"num_tokens": 4116804.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.604962348937988,
|
|
"epoch": 2.0924912978617605,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004836780802164016,
|
|
"loss": 6.4155,
|
|
"mean_token_accuracy": 0.13956694304943085,
|
|
"num_tokens": 4126658.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.560297727584839,
|
|
"epoch": 2.0974639482844357,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048353173643868064,
|
|
"loss": 6.1869,
|
|
"mean_token_accuracy": 0.1467223197221756,
|
|
"num_tokens": 4136323.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.513409566879273,
|
|
"epoch": 2.102436598707111,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004833847644367826,
|
|
"loss": 6.3312,
|
|
"mean_token_accuracy": 0.1473471313714981,
|
|
"num_tokens": 4145714.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.651977109909057,
|
|
"epoch": 2.107409249129786,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048323716465347615,
|
|
"loss": 6.3341,
|
|
"mean_token_accuracy": 0.14381592273712157,
|
|
"num_tokens": 4156059.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.5120772361755375,
|
|
"epoch": 2.1123818995524615,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004830889375334215,
|
|
"loss": 6.3755,
|
|
"mean_token_accuracy": 0.13871776461601257,
|
|
"num_tokens": 4166990.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.6102838039398195,
|
|
"epoch": 2.1173545499751367,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048294008352316875,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.14244372248649598,
|
|
"num_tokens": 4175716.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.610816287994385,
|
|
"epoch": 2.122327200397812,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004827906030711564,
|
|
"loss": 6.357,
|
|
"mean_token_accuracy": 0.1364901088178158,
|
|
"num_tokens": 4184832.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.389426517486572,
|
|
"epoch": 2.1272998508204872,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048264049662771053,
|
|
"loss": 6.1327,
|
|
"mean_token_accuracy": 0.1567293621599674,
|
|
"num_tokens": 4195678.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.552158784866333,
|
|
"epoch": 2.1322725012431625,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048248976464504266,
|
|
"loss": 6.3209,
|
|
"mean_token_accuracy": 0.14384701550006868,
|
|
"num_tokens": 4205357.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.641371536254883,
|
|
"epoch": 2.1372451516658377,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048233840757724926,
|
|
"loss": 6.3945,
|
|
"mean_token_accuracy": 0.13992423713207244,
|
|
"num_tokens": 4215416.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.614249420166016,
|
|
"epoch": 2.142217802088513,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048218642588030943,
|
|
"loss": 6.4327,
|
|
"mean_token_accuracy": 0.1366390034556389,
|
|
"num_tokens": 4225894.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.64572958946228,
|
|
"epoch": 2.1471904525111887,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048203382001208453,
|
|
"loss": 6.358,
|
|
"mean_token_accuracy": 0.1440119169652462,
|
|
"num_tokens": 4235973.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.531086492538452,
|
|
"epoch": 2.152163102933864,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048188059043231585,
|
|
"loss": 6.2773,
|
|
"mean_token_accuracy": 0.14828858226537706,
|
|
"num_tokens": 4246411.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.590059375762939,
|
|
"epoch": 2.157135753356539,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048172673760262414,
|
|
"loss": 6.344,
|
|
"mean_token_accuracy": 0.14175421819090844,
|
|
"num_tokens": 4255754.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.502303123474121,
|
|
"epoch": 2.1621084037792144,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048157226198650735,
|
|
"loss": 6.2693,
|
|
"mean_token_accuracy": 0.14281183630228042,
|
|
"num_tokens": 4265263.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.69334225654602,
|
|
"epoch": 2.1670810542018897,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004814171640493398,
|
|
"loss": 6.4463,
|
|
"mean_token_accuracy": 0.13476933911442757,
|
|
"num_tokens": 4275262.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.518319416046142,
|
|
"epoch": 2.172053704624565,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004812614442583708,
|
|
"loss": 6.358,
|
|
"mean_token_accuracy": 0.14073809012770652,
|
|
"num_tokens": 4285467.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.573318862915039,
|
|
"epoch": 2.17702635504724,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004811051030827227,
|
|
"loss": 6.2853,
|
|
"mean_token_accuracy": 0.13876265659928322,
|
|
"num_tokens": 4294680.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.5653125762939455,
|
|
"epoch": 2.1819990054699154,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048094814099339006,
|
|
"loss": 6.3742,
|
|
"mean_token_accuracy": 0.13634245693683625,
|
|
"num_tokens": 4304605.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.42925910949707,
|
|
"epoch": 2.1869716558925907,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004807905584632381,
|
|
"loss": 6.189,
|
|
"mean_token_accuracy": 0.15406806021928787,
|
|
"num_tokens": 4314112.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.470149326324463,
|
|
"epoch": 2.191944306315266,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048063235596700087,
|
|
"loss": 6.2381,
|
|
"mean_token_accuracy": 0.14347675293684006,
|
|
"num_tokens": 4323872.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.5649285316467285,
|
|
"epoch": 2.196916956737941,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048047353398128056,
|
|
"loss": 6.2145,
|
|
"mean_token_accuracy": 0.1485954388976097,
|
|
"num_tokens": 4333002.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.486789560317993,
|
|
"epoch": 2.2018896071606164,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004803140929845452,
|
|
"loss": 6.3213,
|
|
"mean_token_accuracy": 0.14201780110597612,
|
|
"num_tokens": 4342012.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.563620662689209,
|
|
"epoch": 2.2068622575832917,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048015403345712796,
|
|
"loss": 6.263,
|
|
"mean_token_accuracy": 0.14286984950304032,
|
|
"num_tokens": 4351959.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.599358415603637,
|
|
"epoch": 2.2118349080059674,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004799933558812252,
|
|
"loss": 6.252,
|
|
"mean_token_accuracy": 0.1491381347179413,
|
|
"num_tokens": 4361267.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.463547706604004,
|
|
"epoch": 2.2168075584286426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047983206074089546,
|
|
"loss": 6.3062,
|
|
"mean_token_accuracy": 0.14415338784456252,
|
|
"num_tokens": 4370104.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.212177658081055,
|
|
"epoch": 2.221780208851318,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004796701485220576,
|
|
"loss": 6.0623,
|
|
"mean_token_accuracy": 0.17453712224960327,
|
|
"num_tokens": 4380724.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.373070907592774,
|
|
"epoch": 2.226752859273993,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004795076197124893,
|
|
"loss": 6.1398,
|
|
"mean_token_accuracy": 0.157767191529274,
|
|
"num_tokens": 4390521.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.522678565979004,
|
|
"epoch": 2.2317255096966684,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047934447480182626,
|
|
"loss": 6.3089,
|
|
"mean_token_accuracy": 0.14738831371068956,
|
|
"num_tokens": 4400075.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.4567193508148195,
|
|
"epoch": 2.2366981601193436,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004791807142815598,
|
|
"loss": 6.2507,
|
|
"mean_token_accuracy": 0.14334490597248079,
|
|
"num_tokens": 4409913.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.5294451236724855,
|
|
"epoch": 2.241670810542019,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047901633864503614,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.14623116999864577,
|
|
"num_tokens": 4418551.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.4681761264801025,
|
|
"epoch": 2.246643460964694,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047885134838745433,
|
|
"loss": 6.4739,
|
|
"mean_token_accuracy": 0.13877973556518555,
|
|
"num_tokens": 4428129.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.717575788497925,
|
|
"epoch": 2.2516161113873694,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004786857440058654,
|
|
"loss": 6.3577,
|
|
"mean_token_accuracy": 0.14355020299553872,
|
|
"num_tokens": 4437311.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.377275991439819,
|
|
"epoch": 2.2565887618100446,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004785195259991701,
|
|
"loss": 6.1703,
|
|
"mean_token_accuracy": 0.15661988630890847,
|
|
"num_tokens": 4447387.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.453979444503784,
|
|
"epoch": 2.26156141223272,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000478352694868118,
|
|
"loss": 6.3669,
|
|
"mean_token_accuracy": 0.14046559035778045,
|
|
"num_tokens": 4456435.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.392428874969482,
|
|
"epoch": 2.266534062655395,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004781852511153058,
|
|
"loss": 6.1772,
|
|
"mean_token_accuracy": 0.1567671611905098,
|
|
"num_tokens": 4466464.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.538432216644287,
|
|
"epoch": 2.271506713078071,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047801719524517573,
|
|
"loss": 6.2989,
|
|
"mean_token_accuracy": 0.14424719288945198,
|
|
"num_tokens": 4476793.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.48081693649292,
|
|
"epoch": 2.276479363500746,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047784852776401403,
|
|
"loss": 6.273,
|
|
"mean_token_accuracy": 0.14824249744415283,
|
|
"num_tokens": 4485811.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.507344913482666,
|
|
"epoch": 2.2814520139234213,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004776792491799495,
|
|
"loss": 6.3846,
|
|
"mean_token_accuracy": 0.14298476576805114,
|
|
"num_tokens": 4495018.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.565403079986572,
|
|
"epoch": 2.2864246643460966,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047750936000295213,
|
|
"loss": 6.3551,
|
|
"mean_token_accuracy": 0.13973737508058548,
|
|
"num_tokens": 4505847.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.485449171066284,
|
|
"epoch": 2.291397314768772,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047733886074483115,
|
|
"loss": 6.3665,
|
|
"mean_token_accuracy": 0.14059092104434967,
|
|
"num_tokens": 4516104.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.348464298248291,
|
|
"epoch": 2.296369965191447,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004771677519192339,
|
|
"loss": 6.098,
|
|
"mean_token_accuracy": 0.16406817212700844,
|
|
"num_tokens": 4524875.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.500853824615478,
|
|
"epoch": 2.3013426156141223,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004769960340416439,
|
|
"loss": 6.3615,
|
|
"mean_token_accuracy": 0.1440914198756218,
|
|
"num_tokens": 4533836.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.246683120727539,
|
|
"epoch": 2.3063152660367976,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047682370762937976,
|
|
"loss": 5.8502,
|
|
"mean_token_accuracy": 0.17627975344657898,
|
|
"num_tokens": 4544323.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.465043735504151,
|
|
"epoch": 2.311287916459473,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047665077320159315,
|
|
"loss": 6.2988,
|
|
"mean_token_accuracy": 0.14809127897024155,
|
|
"num_tokens": 4553811.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.340263938903808,
|
|
"epoch": 2.316260566882148,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00047647723127926765,
|
|
"loss": 6.1237,
|
|
"mean_token_accuracy": 0.16093990504741668,
|
|
"num_tokens": 4563098.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.516020965576172,
|
|
"epoch": 2.3212332173048233,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004763030823852168,
|
|
"loss": 6.4184,
|
|
"mean_token_accuracy": 0.14068740233778954,
|
|
"num_tokens": 4573589.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.527627563476562,
|
|
"epoch": 2.3262058677274986,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004761283270440829,
|
|
"loss": 6.3068,
|
|
"mean_token_accuracy": 0.14331134855747224,
|
|
"num_tokens": 4583059.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.597674226760864,
|
|
"epoch": 2.331178518150174,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004759529657823349,
|
|
"loss": 6.478,
|
|
"mean_token_accuracy": 0.13969092965126037,
|
|
"num_tokens": 4593371.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.519469928741455,
|
|
"epoch": 2.336151168572849,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047577699912826755,
|
|
"loss": 6.3905,
|
|
"mean_token_accuracy": 0.13801799044013024,
|
|
"num_tokens": 4602862.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.496824026107788,
|
|
"epoch": 2.341123818995525,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047560042761199915,
|
|
"loss": 6.2651,
|
|
"mean_token_accuracy": 0.14610372334718705,
|
|
"num_tokens": 4612827.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.515990734100342,
|
|
"epoch": 2.3460964694182,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004754232517654704,
|
|
"loss": 6.3304,
|
|
"mean_token_accuracy": 0.1415424481034279,
|
|
"num_tokens": 4621706.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.525302171707153,
|
|
"epoch": 2.3510691198408753,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004752454721224424,
|
|
"loss": 6.2802,
|
|
"mean_token_accuracy": 0.14266075789928437,
|
|
"num_tokens": 4630878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.553720140457154,
|
|
"epoch": 2.3560417702635505,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047506708921849535,
|
|
"loss": 6.4066,
|
|
"mean_token_accuracy": 0.13664366900920868,
|
|
"num_tokens": 4640582.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.473950529098511,
|
|
"epoch": 2.361014420686226,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004748881035910269,
|
|
"loss": 6.3303,
|
|
"mean_token_accuracy": 0.139693483710289,
|
|
"num_tokens": 4650799.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.509372663497925,
|
|
"epoch": 2.365987071108901,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004747085157792504,
|
|
"loss": 6.2819,
|
|
"mean_token_accuracy": 0.14855707809329033,
|
|
"num_tokens": 4660566.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.479830694198609,
|
|
"epoch": 2.3709597215315763,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004745283263241933,
|
|
"loss": 6.3665,
|
|
"mean_token_accuracy": 0.13882315009832383,
|
|
"num_tokens": 4670234.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.527936601638794,
|
|
"epoch": 2.3759323719542516,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004743475357686957,
|
|
"loss": 6.3082,
|
|
"mean_token_accuracy": 0.14590878039598465,
|
|
"num_tokens": 4681114.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.418662023544312,
|
|
"epoch": 2.380905022376927,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047416614465740843,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.17148747891187668,
|
|
"num_tokens": 4691970.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.387086582183838,
|
|
"epoch": 2.385877672799602,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047398415353679167,
|
|
"loss": 6.3107,
|
|
"mean_token_accuracy": 0.14175716415047646,
|
|
"num_tokens": 4701723.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.486026048660278,
|
|
"epoch": 2.3908503232222773,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004738015629551132,
|
|
"loss": 6.2964,
|
|
"mean_token_accuracy": 0.1505607545375824,
|
|
"num_tokens": 4711356.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.527990865707397,
|
|
"epoch": 2.395822973644953,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047361837346244655,
|
|
"loss": 6.3321,
|
|
"mean_token_accuracy": 0.14844110757112502,
|
|
"num_tokens": 4721765.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.456621742248535,
|
|
"epoch": 2.4007956240676283,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004734345856106698,
|
|
"loss": 6.2086,
|
|
"mean_token_accuracy": 0.1463665895164013,
|
|
"num_tokens": 4731678.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.412936639785767,
|
|
"epoch": 2.4057682744903035,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047325019995346355,
|
|
"loss": 6.2297,
|
|
"mean_token_accuracy": 0.14574960023164749,
|
|
"num_tokens": 4740960.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.421316862106323,
|
|
"epoch": 2.4107409249129788,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004730652170463091,
|
|
"loss": 6.3602,
|
|
"mean_token_accuracy": 0.13718757703900336,
|
|
"num_tokens": 4751753.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.50187120437622,
|
|
"epoch": 2.415713575335654,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004728796374464875,
|
|
"loss": 6.2239,
|
|
"mean_token_accuracy": 0.15039970725774765,
|
|
"num_tokens": 4761274.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.451924514770508,
|
|
"epoch": 2.4206862257583293,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047269346171307716,
|
|
"loss": 6.3944,
|
|
"mean_token_accuracy": 0.14462558403611184,
|
|
"num_tokens": 4770852.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.567006349563599,
|
|
"epoch": 2.4256588761810045,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004725066904069523,
|
|
"loss": 6.259,
|
|
"mean_token_accuracy": 0.1490871302783489,
|
|
"num_tokens": 4780857.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.393218326568603,
|
|
"epoch": 2.4306315266036798,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004723193240907814,
|
|
"loss": 6.2177,
|
|
"mean_token_accuracy": 0.14124572947621344,
|
|
"num_tokens": 4790018.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.431617975234985,
|
|
"epoch": 2.435604177026355,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004721313633290257,
|
|
"loss": 6.3018,
|
|
"mean_token_accuracy": 0.14912513345479966,
|
|
"num_tokens": 4799747.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.528079700469971,
|
|
"epoch": 2.4405768274490303,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000471942808687937,
|
|
"loss": 6.3227,
|
|
"mean_token_accuracy": 0.14339413195848466,
|
|
"num_tokens": 4809710.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.447162294387818,
|
|
"epoch": 2.4455494778717055,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047175366073555654,
|
|
"loss": 6.2912,
|
|
"mean_token_accuracy": 0.14973672702908516,
|
|
"num_tokens": 4819387.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.417529726028443,
|
|
"epoch": 2.4505221282943808,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004715639200417125,
|
|
"loss": 6.2294,
|
|
"mean_token_accuracy": 0.14512860402464867,
|
|
"num_tokens": 4829541.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.40480751991272,
|
|
"epoch": 2.455494778717056,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004713735871780192,
|
|
"loss": 6.2187,
|
|
"mean_token_accuracy": 0.1514206737279892,
|
|
"num_tokens": 4838699.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.464950466156006,
|
|
"epoch": 2.4604674291397313,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004711826627178747,
|
|
"loss": 6.3591,
|
|
"mean_token_accuracy": 0.142837306112051,
|
|
"num_tokens": 4848295.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.4626429080963135,
|
|
"epoch": 2.4654400795624065,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047099114723645944,
|
|
"loss": 6.2243,
|
|
"mean_token_accuracy": 0.15163095891475678,
|
|
"num_tokens": 4857649.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.451371669769287,
|
|
"epoch": 2.470412729985082,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047079904131073434,
|
|
"loss": 6.2392,
|
|
"mean_token_accuracy": 0.15429284498095514,
|
|
"num_tokens": 4867332.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.471212100982666,
|
|
"epoch": 2.4753853804077575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004706063455194389,
|
|
"loss": 6.3304,
|
|
"mean_token_accuracy": 0.13392013162374497,
|
|
"num_tokens": 4878726.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.511387825012207,
|
|
"epoch": 2.4803580308304327,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047041306044308986,
|
|
"loss": 6.3205,
|
|
"mean_token_accuracy": 0.14299636483192443,
|
|
"num_tokens": 4888030.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.426170015335083,
|
|
"epoch": 2.485330681253108,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004702191866639792,
|
|
"loss": 6.3084,
|
|
"mean_token_accuracy": 0.1375191852450371,
|
|
"num_tokens": 4897191.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.485330681253108,
|
|
"eval_entropy": 6.344454524703838,
|
|
"eval_loss": 6.799404621124268,
|
|
"eval_mean_token_accuracy": 0.12806895586408779,
|
|
"eval_num_tokens": 4897191.0,
|
|
"eval_runtime": 13.1007,
|
|
"eval_samples_per_second": 264.489,
|
|
"eval_steps_per_second": 33.128,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.473689889907837,
|
|
"epoch": 2.490303331675783,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004700247247661726,
|
|
"loss": 6.2993,
|
|
"mean_token_accuracy": 0.14937970638275147,
|
|
"num_tokens": 4907154.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.2477129936218265,
|
|
"epoch": 2.4952759820984585,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004698296753355072,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.17377749904990197,
|
|
"num_tokens": 4917582.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.390049886703491,
|
|
"epoch": 2.5002486325211337,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004696340389595903,
|
|
"loss": 6.2371,
|
|
"mean_token_accuracy": 0.15294406563043594,
|
|
"num_tokens": 4927473.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.4877838611602785,
|
|
"epoch": 2.505221282943809,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046943781622779733,
|
|
"loss": 6.3658,
|
|
"mean_token_accuracy": 0.1369009792804718,
|
|
"num_tokens": 4938037.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.45825719833374,
|
|
"epoch": 2.510193933366484,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046924100773127055,
|
|
"loss": 6.3182,
|
|
"mean_token_accuracy": 0.13753349408507348,
|
|
"num_tokens": 4948009.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.393920755386352,
|
|
"epoch": 2.5151665837891595,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046904361406291633,
|
|
"loss": 6.284,
|
|
"mean_token_accuracy": 0.1474185824394226,
|
|
"num_tokens": 4957971.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.541047620773315,
|
|
"epoch": 2.520139234211835,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004688456358174044,
|
|
"loss": 6.2853,
|
|
"mean_token_accuracy": 0.13737846910953522,
|
|
"num_tokens": 4968040.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.377184295654297,
|
|
"epoch": 2.5251118846345104,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004686470735911653,
|
|
"loss": 6.2084,
|
|
"mean_token_accuracy": 0.1514485388994217,
|
|
"num_tokens": 4977142.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.4258208751678465,
|
|
"epoch": 2.5300845350571857,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004684479279823891,
|
|
"loss": 6.3988,
|
|
"mean_token_accuracy": 0.14306986778974534,
|
|
"num_tokens": 4986726.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.53545298576355,
|
|
"epoch": 2.535057185479861,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004682481995910232,
|
|
"loss": 6.2778,
|
|
"mean_token_accuracy": 0.14280636683106424,
|
|
"num_tokens": 4996690.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.42930006980896,
|
|
"epoch": 2.540029835902536,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046804788901877084,
|
|
"loss": 6.3299,
|
|
"mean_token_accuracy": 0.14825762957334518,
|
|
"num_tokens": 5005804.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.405391073226928,
|
|
"epoch": 2.5450024863252114,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046784699686908886,
|
|
"loss": 6.2746,
|
|
"mean_token_accuracy": 0.14945762753486633,
|
|
"num_tokens": 5015666.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.40090126991272,
|
|
"epoch": 2.5499751367478867,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046764552374718656,
|
|
"loss": 6.2074,
|
|
"mean_token_accuracy": 0.15531133264303207,
|
|
"num_tokens": 5025897.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.449265956878662,
|
|
"epoch": 2.554947787170562,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046744347026002315,
|
|
"loss": 6.2779,
|
|
"mean_token_accuracy": 0.14622714519500732,
|
|
"num_tokens": 5035195.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.417751693725586,
|
|
"epoch": 2.559920437593237,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004672408370163066,
|
|
"loss": 6.3101,
|
|
"mean_token_accuracy": 0.1482762038707733,
|
|
"num_tokens": 5044472.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.408794212341308,
|
|
"epoch": 2.5648930880159124,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000467037624626491,
|
|
"loss": 6.3535,
|
|
"mean_token_accuracy": 0.1427104167640209,
|
|
"num_tokens": 5053593.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.44328761100769,
|
|
"epoch": 2.5698657384385877,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046683383370277554,
|
|
"loss": 6.2976,
|
|
"mean_token_accuracy": 0.14150633215904235,
|
|
"num_tokens": 5063604.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.424183797836304,
|
|
"epoch": 2.574838388861263,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046662946485910226,
|
|
"loss": 6.2005,
|
|
"mean_token_accuracy": 0.14552666246891022,
|
|
"num_tokens": 5073013.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.368274736404419,
|
|
"epoch": 2.579811039283938,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000466424518711154,
|
|
"loss": 6.2824,
|
|
"mean_token_accuracy": 0.14596861377358436,
|
|
"num_tokens": 5082590.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.500651693344116,
|
|
"epoch": 2.5847836897066134,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004662189958763531,
|
|
"loss": 6.2653,
|
|
"mean_token_accuracy": 0.1461903840303421,
|
|
"num_tokens": 5091860.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.410395240783691,
|
|
"epoch": 2.5897563401292887,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004660128969738591,
|
|
"loss": 6.1871,
|
|
"mean_token_accuracy": 0.15528345704078675,
|
|
"num_tokens": 5101164.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.293390130996704,
|
|
"epoch": 2.594728990551964,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004658062226245668,
|
|
"loss": 6.2705,
|
|
"mean_token_accuracy": 0.1464535154402256,
|
|
"num_tokens": 5112885.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.456001424789429,
|
|
"epoch": 2.5997016409746396,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004655989734511052,
|
|
"loss": 6.2578,
|
|
"mean_token_accuracy": 0.15007225126028062,
|
|
"num_tokens": 5124063.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.287815856933594,
|
|
"epoch": 2.604674291397315,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004653911500778343,
|
|
"loss": 6.0849,
|
|
"mean_token_accuracy": 0.16392876207828522,
|
|
"num_tokens": 5134286.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.430450820922852,
|
|
"epoch": 2.60964694181999,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004651827531308443,
|
|
"loss": 6.3578,
|
|
"mean_token_accuracy": 0.14109658673405648,
|
|
"num_tokens": 5144421.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.532381200790406,
|
|
"epoch": 2.6146195922426654,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004649737832379534,
|
|
"loss": 6.2944,
|
|
"mean_token_accuracy": 0.14692365378141403,
|
|
"num_tokens": 5153466.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.357282972335815,
|
|
"epoch": 2.6195922426653406,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004647642410287057,
|
|
"loss": 6.3585,
|
|
"mean_token_accuracy": 0.1402343362569809,
|
|
"num_tokens": 5164034.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.389232969284057,
|
|
"epoch": 2.624564893088016,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004645541271343696,
|
|
"loss": 6.2871,
|
|
"mean_token_accuracy": 0.14616041854023934,
|
|
"num_tokens": 5173867.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.463972425460815,
|
|
"epoch": 2.629537543510691,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046434344218793566,
|
|
"loss": 6.3267,
|
|
"mean_token_accuracy": 0.14457016289234162,
|
|
"num_tokens": 5183182.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.408384943008423,
|
|
"epoch": 2.6345101939333664,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046413218682411485,
|
|
"loss": 6.2947,
|
|
"mean_token_accuracy": 0.14783588126301767,
|
|
"num_tokens": 5192537.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.465983772277832,
|
|
"epoch": 2.6394828443560416,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004639203616793365,
|
|
"loss": 6.2529,
|
|
"mean_token_accuracy": 0.15335736125707627,
|
|
"num_tokens": 5201879.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.43082480430603,
|
|
"epoch": 2.6444554947787173,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046370796739174657,
|
|
"loss": 6.2722,
|
|
"mean_token_accuracy": 0.14541275426745415,
|
|
"num_tokens": 5211555.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.369832277297974,
|
|
"epoch": 2.6494281452013926,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046349500460120556,
|
|
"loss": 6.1719,
|
|
"mean_token_accuracy": 0.1572995215654373,
|
|
"num_tokens": 5220895.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.455500268936158,
|
|
"epoch": 2.654400795624068,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046328147394928684,
|
|
"loss": 6.3361,
|
|
"mean_token_accuracy": 0.1436840482056141,
|
|
"num_tokens": 5230101.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.483569717407226,
|
|
"epoch": 2.659373446046743,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004630673760792741,
|
|
"loss": 6.3253,
|
|
"mean_token_accuracy": 0.1354352928698063,
|
|
"num_tokens": 5239472.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.328264045715332,
|
|
"epoch": 2.6643460964694183,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004628527116361603,
|
|
"loss": 6.2573,
|
|
"mean_token_accuracy": 0.1515199825167656,
|
|
"num_tokens": 5248823.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.398807668685913,
|
|
"epoch": 2.6693187468920936,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004626374812666449,
|
|
"loss": 6.278,
|
|
"mean_token_accuracy": 0.14263736754655837,
|
|
"num_tokens": 5258551.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.396531295776367,
|
|
"epoch": 2.674291397314769,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046242168561913256,
|
|
"loss": 6.0728,
|
|
"mean_token_accuracy": 0.16309110224246978,
|
|
"num_tokens": 5267933.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.442751359939575,
|
|
"epoch": 2.679264047737444,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046220532534373067,
|
|
"loss": 6.3232,
|
|
"mean_token_accuracy": 0.14714367985725402,
|
|
"num_tokens": 5277856.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.458986473083496,
|
|
"epoch": 2.6842366981601193,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004619884010922477,
|
|
"loss": 6.3166,
|
|
"mean_token_accuracy": 0.138733346760273,
|
|
"num_tokens": 5288127.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.410635948181152,
|
|
"epoch": 2.6892093485827946,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004617709135181913,
|
|
"loss": 6.2652,
|
|
"mean_token_accuracy": 0.14450549632310866,
|
|
"num_tokens": 5297143.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.3663300514221195,
|
|
"epoch": 2.69418199900547,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00046155286327676594,
|
|
"loss": 6.2413,
|
|
"mean_token_accuracy": 0.14712165296077728,
|
|
"num_tokens": 5307067.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.359155893325806,
|
|
"epoch": 2.699154649428145,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046133425102487134,
|
|
"loss": 6.2027,
|
|
"mean_token_accuracy": 0.15176677256822585,
|
|
"num_tokens": 5317471.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.480701446533203,
|
|
"epoch": 2.7041272998508203,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046111507742110035,
|
|
"loss": 6.2613,
|
|
"mean_token_accuracy": 0.15174250081181526,
|
|
"num_tokens": 5326803.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.379540014266968,
|
|
"epoch": 2.7090999502734956,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000460895343125737,
|
|
"loss": 6.2512,
|
|
"mean_token_accuracy": 0.15086393505334855,
|
|
"num_tokens": 5336342.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.206189298629761,
|
|
"epoch": 2.714072600696171,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046067504880075425,
|
|
"loss": 6.0802,
|
|
"mean_token_accuracy": 0.17169879078865052,
|
|
"num_tokens": 5345694.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.445157194137574,
|
|
"epoch": 2.719045251118846,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004604541951098122,
|
|
"loss": 6.1626,
|
|
"mean_token_accuracy": 0.1573153018951416,
|
|
"num_tokens": 5355277.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.375748777389527,
|
|
"epoch": 2.724017901541522,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004602327827182565,
|
|
"loss": 6.2097,
|
|
"mean_token_accuracy": 0.15330390483140946,
|
|
"num_tokens": 5364650.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.35117654800415,
|
|
"epoch": 2.728990551964197,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004600108122931156,
|
|
"loss": 6.185,
|
|
"mean_token_accuracy": 0.148213666677475,
|
|
"num_tokens": 5373151.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.378377151489258,
|
|
"epoch": 2.7339632023868723,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045978828450309907,
|
|
"loss": 6.3039,
|
|
"mean_token_accuracy": 0.148667760938406,
|
|
"num_tokens": 5383093.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.4949675559997555,
|
|
"epoch": 2.7389358528095475,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.0004595652000185958,
|
|
"loss": 6.1987,
|
|
"mean_token_accuracy": 0.15289298444986343,
|
|
"num_tokens": 5392060.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.400454711914063,
|
|
"epoch": 2.743908503232223,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045934155951167163,
|
|
"loss": 6.3268,
|
|
"mean_token_accuracy": 0.14166544750332832,
|
|
"num_tokens": 5402119.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.395109510421753,
|
|
"epoch": 2.748881153654898,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004591173636560675,
|
|
"loss": 6.2402,
|
|
"mean_token_accuracy": 0.15237922072410584,
|
|
"num_tokens": 5411804.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.3739283084869385,
|
|
"epoch": 2.7538538040775733,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004588926131271974,
|
|
"loss": 6.2926,
|
|
"mean_token_accuracy": 0.14396035075187683,
|
|
"num_tokens": 5423808.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.417657947540283,
|
|
"epoch": 2.7588264545002485,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045866730860214655,
|
|
"loss": 6.2457,
|
|
"mean_token_accuracy": 0.14892006367444993,
|
|
"num_tokens": 5432154.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.224423599243164,
|
|
"epoch": 2.763799104922924,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004584414507596687,
|
|
"loss": 6.0062,
|
|
"mean_token_accuracy": 0.16566762328147888,
|
|
"num_tokens": 5443006.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.330063533782959,
|
|
"epoch": 2.768771755345599,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045821504028018494,
|
|
"loss": 6.1658,
|
|
"mean_token_accuracy": 0.1499434307217598,
|
|
"num_tokens": 5453596.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.4482732772827145,
|
|
"epoch": 2.7737444057682747,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004579880778457811,
|
|
"loss": 6.2919,
|
|
"mean_token_accuracy": 0.14520807862281798,
|
|
"num_tokens": 5463048.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.3992150783538815,
|
|
"epoch": 2.77871705619095,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004577605641402056,
|
|
"loss": 6.2377,
|
|
"mean_token_accuracy": 0.15053599774837495,
|
|
"num_tokens": 5472409.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.3354569435119625,
|
|
"epoch": 2.7836897066136252,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004575324998488681,
|
|
"loss": 6.169,
|
|
"mean_token_accuracy": 0.15516238659620285,
|
|
"num_tokens": 5481380.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.398344039916992,
|
|
"epoch": 2.7886623570363005,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004573038856588364,
|
|
"loss": 6.3817,
|
|
"mean_token_accuracy": 0.13996670097112657,
|
|
"num_tokens": 5491157.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.410790014266968,
|
|
"epoch": 2.7936350074589757,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004570747222588354,
|
|
"loss": 6.2644,
|
|
"mean_token_accuracy": 0.14915478974580765,
|
|
"num_tokens": 5500856.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.38122296333313,
|
|
"epoch": 2.798607657881651,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045684501033924443,
|
|
"loss": 6.1437,
|
|
"mean_token_accuracy": 0.15261420160531997,
|
|
"num_tokens": 5509654.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.349528360366821,
|
|
"epoch": 2.8035803083043263,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045661475059209503,
|
|
"loss": 6.1576,
|
|
"mean_token_accuracy": 0.15539332628250122,
|
|
"num_tokens": 5519009.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.287224531173706,
|
|
"epoch": 2.8085529587270015,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004563839437110695,
|
|
"loss": 6.1577,
|
|
"mean_token_accuracy": 0.1548203259706497,
|
|
"num_tokens": 5528316.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.399583053588867,
|
|
"epoch": 2.8135256091496768,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045615259039149826,
|
|
"loss": 6.2897,
|
|
"mean_token_accuracy": 0.15138771012425423,
|
|
"num_tokens": 5538416.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.350771522521972,
|
|
"epoch": 2.818498259572352,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045592069133035783,
|
|
"loss": 6.2494,
|
|
"mean_token_accuracy": 0.14774278923869133,
|
|
"num_tokens": 5548600.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.412031412124634,
|
|
"epoch": 2.8234709099950273,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004556882472262691,
|
|
"loss": 6.3321,
|
|
"mean_token_accuracy": 0.14426151514053345,
|
|
"num_tokens": 5559454.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.400199604034424,
|
|
"epoch": 2.8284435604177025,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004554552587794948,
|
|
"loss": 6.2507,
|
|
"mean_token_accuracy": 0.14439518600702286,
|
|
"num_tokens": 5567748.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.4188232421875,
|
|
"epoch": 2.8334162108403778,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004552217266919375,
|
|
"loss": 6.3639,
|
|
"mean_token_accuracy": 0.13834651559591293,
|
|
"num_tokens": 5577617.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.443787527084351,
|
|
"epoch": 2.838388861263053,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004549876516671376,
|
|
"loss": 6.3255,
|
|
"mean_token_accuracy": 0.14910750165581704,
|
|
"num_tokens": 5586600.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.437282705307007,
|
|
"epoch": 2.8433615116857283,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045475303441027114,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.1601414605975151,
|
|
"num_tokens": 5596526.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.366176795959473,
|
|
"epoch": 2.8483341621084035,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004545178756281479,
|
|
"loss": 6.2199,
|
|
"mean_token_accuracy": 0.15804926306009293,
|
|
"num_tokens": 5605780.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.298853063583374,
|
|
"epoch": 2.853306812531079,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045428217602920876,
|
|
"loss": 6.2716,
|
|
"mean_token_accuracy": 0.14463355615735055,
|
|
"num_tokens": 5616109.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.356553077697754,
|
|
"epoch": 2.8582794629537545,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045404593632352385,
|
|
"loss": 6.2193,
|
|
"mean_token_accuracy": 0.15245952904224397,
|
|
"num_tokens": 5625423.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.4199056148529055,
|
|
"epoch": 2.8632521133764297,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004538091572227909,
|
|
"loss": 6.3047,
|
|
"mean_token_accuracy": 0.14840330481529235,
|
|
"num_tokens": 5634538.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.283694839477539,
|
|
"epoch": 2.868224763799105,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004535718394403319,
|
|
"loss": 6.0773,
|
|
"mean_token_accuracy": 0.16899314150214195,
|
|
"num_tokens": 5644087.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.287930345535278,
|
|
"epoch": 2.87319741422178,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004533339836910924,
|
|
"loss": 6.1231,
|
|
"mean_token_accuracy": 0.15990129113197327,
|
|
"num_tokens": 5654241.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.312622594833374,
|
|
"epoch": 2.8781700646444555,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004530955906916382,
|
|
"loss": 6.1734,
|
|
"mean_token_accuracy": 0.15418502539396287,
|
|
"num_tokens": 5663827.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.366262626647949,
|
|
"epoch": 2.8831427150671307,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004528566611601536,
|
|
"loss": 6.2448,
|
|
"mean_token_accuracy": 0.1495486930012703,
|
|
"num_tokens": 5673427.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.352624368667603,
|
|
"epoch": 2.888115365489806,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004526171958164396,
|
|
"loss": 6.1485,
|
|
"mean_token_accuracy": 0.1462064042687416,
|
|
"num_tokens": 5684762.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.237772703170776,
|
|
"epoch": 2.893088015912481,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045237719538191107,
|
|
"loss": 6.2351,
|
|
"mean_token_accuracy": 0.15170298516750336,
|
|
"num_tokens": 5694938.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.335097932815552,
|
|
"epoch": 2.898060666335157,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000452136660579595,
|
|
"loss": 6.1486,
|
|
"mean_token_accuracy": 0.15374543964862825,
|
|
"num_tokens": 5704910.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.410652446746826,
|
|
"epoch": 2.903033316757832,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045189559213412825,
|
|
"loss": 6.2315,
|
|
"mean_token_accuracy": 0.15082828253507613,
|
|
"num_tokens": 5714325.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.295235300064087,
|
|
"epoch": 2.9080059671805074,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004516539907717553,
|
|
"loss": 6.1726,
|
|
"mean_token_accuracy": 0.15955992937088012,
|
|
"num_tokens": 5724382.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.331395196914673,
|
|
"epoch": 2.9129786176031827,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045141185722032617,
|
|
"loss": 6.2054,
|
|
"mean_token_accuracy": 0.15302468985319137,
|
|
"num_tokens": 5734277.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.4001991748809814,
|
|
"epoch": 2.917951268025858,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045116919220929407,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.14625603780150415,
|
|
"num_tokens": 5742772.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.290955305099487,
|
|
"epoch": 2.922923918448533,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004509259964697133,
|
|
"loss": 6.231,
|
|
"mean_token_accuracy": 0.14993418306112288,
|
|
"num_tokens": 5754011.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.487302350997925,
|
|
"epoch": 2.9278965688712084,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000450682270734237,
|
|
"loss": 6.3289,
|
|
"mean_token_accuracy": 0.14362795278429985,
|
|
"num_tokens": 5765445.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.364766645431518,
|
|
"epoch": 2.9328692192938837,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004504380157371152,
|
|
"loss": 6.2485,
|
|
"mean_token_accuracy": 0.1522974729537964,
|
|
"num_tokens": 5775529.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.372902202606201,
|
|
"epoch": 2.937841869716559,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045019323221419204,
|
|
"loss": 6.3711,
|
|
"mean_token_accuracy": 0.14461278319358825,
|
|
"num_tokens": 5785594.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.360442686080932,
|
|
"epoch": 2.942814520139234,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044994792090290425,
|
|
"loss": 6.0868,
|
|
"mean_token_accuracy": 0.16279054135084153,
|
|
"num_tokens": 5796071.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.299129247665405,
|
|
"epoch": 2.9477871705619094,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044970208254227836,
|
|
"loss": 6.2097,
|
|
"mean_token_accuracy": 0.15824256241321563,
|
|
"num_tokens": 5805127.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.412131071090698,
|
|
"epoch": 2.9527598209845847,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044945571787292866,
|
|
"loss": 6.2866,
|
|
"mean_token_accuracy": 0.14842584431171418,
|
|
"num_tokens": 5815001.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.361129379272461,
|
|
"epoch": 2.95773247140726,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004492088276370552,
|
|
"loss": 6.2296,
|
|
"mean_token_accuracy": 0.14568739533424377,
|
|
"num_tokens": 5826915.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.362636470794678,
|
|
"epoch": 2.962705121829935,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004489614125784411,
|
|
"loss": 6.2258,
|
|
"mean_token_accuracy": 0.1487402908504009,
|
|
"num_tokens": 5837444.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.316435384750366,
|
|
"epoch": 2.9676777722526104,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004487134734424509,
|
|
"loss": 6.0954,
|
|
"mean_token_accuracy": 0.15558630004525184,
|
|
"num_tokens": 5846514.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.396707201004029,
|
|
"epoch": 2.9726504226752857,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004484650109760276,
|
|
"loss": 6.3707,
|
|
"mean_token_accuracy": 0.1426582582294941,
|
|
"num_tokens": 5857227.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.414143323898315,
|
|
"epoch": 2.9776230730979614,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004482160259276911,
|
|
"loss": 6.3532,
|
|
"mean_token_accuracy": 0.14733145982027054,
|
|
"num_tokens": 5866877.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.332798194885254,
|
|
"epoch": 2.9825957235206366,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004479665190475355,
|
|
"loss": 6.2325,
|
|
"mean_token_accuracy": 0.1496262565255165,
|
|
"num_tokens": 5877127.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.9825957235206366,
|
|
"eval_entropy": 6.238766670227051,
|
|
"eval_loss": 6.684376239776611,
|
|
"eval_mean_token_accuracy": 0.13300726722393716,
|
|
"eval_num_tokens": 5877127.0,
|
|
"eval_runtime": 9.6447,
|
|
"eval_samples_per_second": 359.263,
|
|
"eval_steps_per_second": 44.999,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.336135149002075,
|
|
"epoch": 2.987568373943312,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044771649108722695,
|
|
"loss": 6.2619,
|
|
"mean_token_accuracy": 0.14929309189319612,
|
|
"num_tokens": 5886434.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.331879949569702,
|
|
"epoch": 2.992541024365987,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004474659428000014,
|
|
"loss": 6.2448,
|
|
"mean_token_accuracy": 0.15186997652053832,
|
|
"num_tokens": 5895462.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.4163923263549805,
|
|
"epoch": 2.9975136747886624,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044721487494066254,
|
|
"loss": 6.1948,
|
|
"mean_token_accuracy": 0.15456721782684327,
|
|
"num_tokens": 5905274.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.278531710306804,
|
|
"epoch": 3.00198906016907,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000446963288265579,
|
|
"loss": 5.9353,
|
|
"mean_token_accuracy": 0.1614049126704534,
|
|
"num_tokens": 5914164.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.299293947219849,
|
|
"epoch": 3.0069617105917454,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044671118353268257,
|
|
"loss": 5.7962,
|
|
"mean_token_accuracy": 0.16236044019460677,
|
|
"num_tokens": 5923159.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.2619325637817385,
|
|
"epoch": 3.0119343610144207,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044645856150146593,
|
|
"loss": 5.7618,
|
|
"mean_token_accuracy": 0.17029790878295897,
|
|
"num_tokens": 5932708.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.2763337135314945,
|
|
"epoch": 3.016907011437096,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044620542293297974,
|
|
"loss": 5.7846,
|
|
"mean_token_accuracy": 0.16297874301671983,
|
|
"num_tokens": 5941370.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.28137354850769,
|
|
"epoch": 3.021879661859771,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004459517685898312,
|
|
"loss": 5.8286,
|
|
"mean_token_accuracy": 0.17183317691087724,
|
|
"num_tokens": 5950567.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.326422500610351,
|
|
"epoch": 3.0268523122824464,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044569759923618113,
|
|
"loss": 5.8881,
|
|
"mean_token_accuracy": 0.16437803506851195,
|
|
"num_tokens": 5959242.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.311964702606201,
|
|
"epoch": 3.0318249627051217,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004454429156377419,
|
|
"loss": 5.8926,
|
|
"mean_token_accuracy": 0.1651277408003807,
|
|
"num_tokens": 5968365.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.248036193847656,
|
|
"epoch": 3.036797613127797,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004451877185617751,
|
|
"loss": 5.8636,
|
|
"mean_token_accuracy": 0.15691922903060912,
|
|
"num_tokens": 5978407.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.122200918197632,
|
|
"epoch": 3.041770263550472,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004449320087770894,
|
|
"loss": 5.6024,
|
|
"mean_token_accuracy": 0.18556191697716712,
|
|
"num_tokens": 5988865.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.256516408920288,
|
|
"epoch": 3.046742913973148,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004446757870540376,
|
|
"loss": 5.7635,
|
|
"mean_token_accuracy": 0.17365970015525817,
|
|
"num_tokens": 5998873.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.233163642883301,
|
|
"epoch": 3.051715564395823,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004444190541645153,
|
|
"loss": 5.8059,
|
|
"mean_token_accuracy": 0.16390766203403473,
|
|
"num_tokens": 6007021.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.140491390228272,
|
|
"epoch": 3.0566882148184984,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00044416181088195776,
|
|
"loss": 5.6268,
|
|
"mean_token_accuracy": 0.17477199882268907,
|
|
"num_tokens": 6015884.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.19992961883545,
|
|
"epoch": 3.0616608652411736,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004439040579813379,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.15730293095111847,
|
|
"num_tokens": 6026110.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.089462423324585,
|
|
"epoch": 3.066633515663849,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044364579623916385,
|
|
"loss": 5.6511,
|
|
"mean_token_accuracy": 0.18649205267429353,
|
|
"num_tokens": 6037479.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.226496028900146,
|
|
"epoch": 3.071606166086524,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004433870264334767,
|
|
"loss": 5.8024,
|
|
"mean_token_accuracy": 0.16192902028560638,
|
|
"num_tokens": 6046743.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.219913101196289,
|
|
"epoch": 3.0765788165091994,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004431277493438484,
|
|
"loss": 5.8909,
|
|
"mean_token_accuracy": 0.16141575425863267,
|
|
"num_tokens": 6056982.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.238383960723877,
|
|
"epoch": 3.0815514669318746,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004428679657513789,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.1644573375582695,
|
|
"num_tokens": 6066016.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.102771949768067,
|
|
"epoch": 3.08652411735455,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004426076764386939,
|
|
"loss": 5.7897,
|
|
"mean_token_accuracy": 0.16131604909896852,
|
|
"num_tokens": 6075149.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.284079360961914,
|
|
"epoch": 3.091496767777225,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044234688218994293,
|
|
"loss": 5.7537,
|
|
"mean_token_accuracy": 0.17464680820703507,
|
|
"num_tokens": 6085736.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.1213531494140625,
|
|
"epoch": 3.0964694181999004,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004420855837907967,
|
|
"loss": 5.6765,
|
|
"mean_token_accuracy": 0.1796969547867775,
|
|
"num_tokens": 6096557.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.291248893737793,
|
|
"epoch": 3.1014420686225757,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044182378202844427,
|
|
"loss": 5.9986,
|
|
"mean_token_accuracy": 0.15412187427282334,
|
|
"num_tokens": 6105735.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.312610769271851,
|
|
"epoch": 3.106414719045251,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044156147769159183,
|
|
"loss": 5.8453,
|
|
"mean_token_accuracy": 0.16719935089349747,
|
|
"num_tokens": 6114132.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.213172769546508,
|
|
"epoch": 3.1113873694679266,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004412986715704591,
|
|
"loss": 5.8472,
|
|
"mean_token_accuracy": 0.16057832092046737,
|
|
"num_tokens": 6124341.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.216770648956299,
|
|
"epoch": 3.116360019890602,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004410353644567777,
|
|
"loss": 5.7502,
|
|
"mean_token_accuracy": 0.170139579474926,
|
|
"num_tokens": 6134418.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.193951034545899,
|
|
"epoch": 3.121332670313277,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044077155714378836,
|
|
"loss": 5.8302,
|
|
"mean_token_accuracy": 0.16187757700681688,
|
|
"num_tokens": 6143659.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.2291477680206295,
|
|
"epoch": 3.1263053207359524,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004405072504262391,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.1614644020795822,
|
|
"num_tokens": 6152706.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.27928991317749,
|
|
"epoch": 3.1312779711586276,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000440242445100382,
|
|
"loss": 5.904,
|
|
"mean_token_accuracy": 0.15977407544851302,
|
|
"num_tokens": 6161390.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.246874141693115,
|
|
"epoch": 3.136250621581303,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043997714196397143,
|
|
"loss": 5.8299,
|
|
"mean_token_accuracy": 0.16721719354391099,
|
|
"num_tokens": 6170982.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.280976438522339,
|
|
"epoch": 3.141223272003978,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043971134181626173,
|
|
"loss": 5.9197,
|
|
"mean_token_accuracy": 0.15720055252313614,
|
|
"num_tokens": 6179356.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.2509472370147705,
|
|
"epoch": 3.1461959224266534,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043944504545800396,
|
|
"loss": 5.8591,
|
|
"mean_token_accuracy": 0.16217541992664336,
|
|
"num_tokens": 6188270.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.2280291557312015,
|
|
"epoch": 3.1511685728493286,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004391782536914447,
|
|
"loss": 5.7916,
|
|
"mean_token_accuracy": 0.16598115786910056,
|
|
"num_tokens": 6197624.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.230363273620606,
|
|
"epoch": 3.156141223272004,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043891096732032257,
|
|
"loss": 5.852,
|
|
"mean_token_accuracy": 0.1616559386253357,
|
|
"num_tokens": 6207174.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.083816194534302,
|
|
"epoch": 3.161113873694679,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043864318714986625,
|
|
"loss": 5.6913,
|
|
"mean_token_accuracy": 0.17798609435558319,
|
|
"num_tokens": 6217939.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.198824834823609,
|
|
"epoch": 3.1660865241173544,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043837491398679234,
|
|
"loss": 5.8654,
|
|
"mean_token_accuracy": 0.15919267386198044,
|
|
"num_tokens": 6227857.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.152533435821534,
|
|
"epoch": 3.17105917454003,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004381061486393025,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.15415869653224945,
|
|
"num_tokens": 6236832.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.21399245262146,
|
|
"epoch": 3.1760318249627053,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004378368919170809,
|
|
"loss": 5.8397,
|
|
"mean_token_accuracy": 0.16585760414600373,
|
|
"num_tokens": 6245879.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.222669839859009,
|
|
"epoch": 3.1810044753853806,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043756714463129244,
|
|
"loss": 5.8525,
|
|
"mean_token_accuracy": 0.15758212134242058,
|
|
"num_tokens": 6256350.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.103539562225341,
|
|
"epoch": 3.185977125808056,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004372969075945796,
|
|
"loss": 5.6475,
|
|
"mean_token_accuracy": 0.18329446464776994,
|
|
"num_tokens": 6266085.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.227236032485962,
|
|
"epoch": 3.190949776230731,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004370261816210604,
|
|
"loss": 5.9125,
|
|
"mean_token_accuracy": 0.16186602115631105,
|
|
"num_tokens": 6276499.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.230378723144531,
|
|
"epoch": 3.1959224266534063,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004367549675263258,
|
|
"loss": 5.9015,
|
|
"mean_token_accuracy": 0.1589238703250885,
|
|
"num_tokens": 6286630.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.1736596584320065,
|
|
"epoch": 3.2008950770760816,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004364832661274373,
|
|
"loss": 5.859,
|
|
"mean_token_accuracy": 0.16225760877132417,
|
|
"num_tokens": 6295907.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.192207288742066,
|
|
"epoch": 3.205867727498757,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043621107824292444,
|
|
"loss": 5.7908,
|
|
"mean_token_accuracy": 0.16480867862701415,
|
|
"num_tokens": 6304304.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.1291985511779785,
|
|
"epoch": 3.210840377921432,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004359384046927824,
|
|
"loss": 5.8846,
|
|
"mean_token_accuracy": 0.15800584703683854,
|
|
"num_tokens": 6315041.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.22709641456604,
|
|
"epoch": 3.2158130283441073,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004356652462984693,
|
|
"loss": 5.7791,
|
|
"mean_token_accuracy": 0.16465489864349364,
|
|
"num_tokens": 6324551.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.206820726394653,
|
|
"epoch": 3.2207856787667826,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043539160388290416,
|
|
"loss": 5.8416,
|
|
"mean_token_accuracy": 0.170079205930233,
|
|
"num_tokens": 6335512.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.172986364364624,
|
|
"epoch": 3.225758329189458,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004351174782704641,
|
|
"loss": 5.899,
|
|
"mean_token_accuracy": 0.15837073996663092,
|
|
"num_tokens": 6345987.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.077767658233642,
|
|
"epoch": 3.230730979612133,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043484287028698176,
|
|
"loss": 5.6793,
|
|
"mean_token_accuracy": 0.18761736303567886,
|
|
"num_tokens": 6357617.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.104698610305786,
|
|
"epoch": 3.2357036300348083,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043456778075974306,
|
|
"loss": 5.782,
|
|
"mean_token_accuracy": 0.17218870371580125,
|
|
"num_tokens": 6366123.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.1479579448699955,
|
|
"epoch": 3.240676280457484,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043429221051748487,
|
|
"loss": 5.8922,
|
|
"mean_token_accuracy": 0.1586204007267952,
|
|
"num_tokens": 6375776.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.316050338745117,
|
|
"epoch": 3.2456489308801593,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043401616039039174,
|
|
"loss": 5.9332,
|
|
"mean_token_accuracy": 0.1481866642832756,
|
|
"num_tokens": 6386235.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.24851713180542,
|
|
"epoch": 3.2506215813028345,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004337396312100944,
|
|
"loss": 5.8943,
|
|
"mean_token_accuracy": 0.1592388369143009,
|
|
"num_tokens": 6396919.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.203681898117066,
|
|
"epoch": 3.2555942317255098,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004334626238096665,
|
|
"loss": 5.8743,
|
|
"mean_token_accuracy": 0.16365359276533126,
|
|
"num_tokens": 6406769.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.1601193904876705,
|
|
"epoch": 3.260566882148185,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004331851390236227,
|
|
"loss": 5.8691,
|
|
"mean_token_accuracy": 0.17016182243824005,
|
|
"num_tokens": 6416208.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.186368656158447,
|
|
"epoch": 3.2655395325708603,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043290717768791533,
|
|
"loss": 5.8243,
|
|
"mean_token_accuracy": 0.16431565433740616,
|
|
"num_tokens": 6426125.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.175332117080688,
|
|
"epoch": 3.2705121829935355,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043262874063993285,
|
|
"loss": 5.814,
|
|
"mean_token_accuracy": 0.15584498643875122,
|
|
"num_tokens": 6436671.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.193968677520752,
|
|
"epoch": 3.2754848334162108,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004323498287184967,
|
|
"loss": 5.9087,
|
|
"mean_token_accuracy": 0.15716016292572021,
|
|
"num_tokens": 6447396.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.157467555999756,
|
|
"epoch": 3.280457483838886,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043207044276385884,
|
|
"loss": 5.8416,
|
|
"mean_token_accuracy": 0.16133513897657395,
|
|
"num_tokens": 6457045.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.116765975952148,
|
|
"epoch": 3.2854301342615613,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043179058361769924,
|
|
"loss": 5.881,
|
|
"mean_token_accuracy": 0.16476525962352753,
|
|
"num_tokens": 6466741.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.010049152374267,
|
|
"epoch": 3.2904027846842365,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004315102521231237,
|
|
"loss": 5.6079,
|
|
"mean_token_accuracy": 0.1854873165488243,
|
|
"num_tokens": 6477769.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.2104979991912845,
|
|
"epoch": 3.295375435106912,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004312294491246609,
|
|
"loss": 5.968,
|
|
"mean_token_accuracy": 0.1543874606490135,
|
|
"num_tokens": 6486991.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.2382039546966555,
|
|
"epoch": 3.3003480855295875,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004309481754682597,
|
|
"loss": 5.8514,
|
|
"mean_token_accuracy": 0.15951886773109436,
|
|
"num_tokens": 6496772.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.146762561798096,
|
|
"epoch": 3.3053207359522627,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043066643200128724,
|
|
"loss": 5.8965,
|
|
"mean_token_accuracy": 0.15671370923519135,
|
|
"num_tokens": 6507577.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.185708141326904,
|
|
"epoch": 3.310293386374938,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004303842195725259,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.15967205613851548,
|
|
"num_tokens": 6518040.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.210595893859863,
|
|
"epoch": 3.3152660367976132,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043010153903217074,
|
|
"loss": 5.8485,
|
|
"mean_token_accuracy": 0.16010092049837113,
|
|
"num_tokens": 6529261.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.142863512039185,
|
|
"epoch": 3.3202386872202885,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004298183912318273,
|
|
"loss": 5.754,
|
|
"mean_token_accuracy": 0.16576443910598754,
|
|
"num_tokens": 6537527.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.165848684310913,
|
|
"epoch": 3.3252113376429637,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042953477702450845,
|
|
"loss": 5.8677,
|
|
"mean_token_accuracy": 0.16407923996448517,
|
|
"num_tokens": 6547205.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.20128436088562,
|
|
"epoch": 3.330183988065639,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004292506972646324,
|
|
"loss": 5.9562,
|
|
"mean_token_accuracy": 0.15806340873241426,
|
|
"num_tokens": 6556328.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.200141191482544,
|
|
"epoch": 3.3351566384883142,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042896615280802006,
|
|
"loss": 5.9082,
|
|
"mean_token_accuracy": 0.16246612891554832,
|
|
"num_tokens": 6566414.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.043509817123413,
|
|
"epoch": 3.3401292889109895,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042868114451189164,
|
|
"loss": 5.7263,
|
|
"mean_token_accuracy": 0.18509557843208313,
|
|
"num_tokens": 6577365.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.212612867355347,
|
|
"epoch": 3.3451019393336647,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004283956732348655,
|
|
"loss": 5.82,
|
|
"mean_token_accuracy": 0.16337819695472716,
|
|
"num_tokens": 6586647.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.19574933052063,
|
|
"epoch": 3.35007458975634,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00042810973983695426,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.16156875863671302,
|
|
"num_tokens": 6596955.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.21228609085083,
|
|
"epoch": 3.3550472401790152,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004278233451795629,
|
|
"loss": 5.8105,
|
|
"mean_token_accuracy": 0.1681644067168236,
|
|
"num_tokens": 6606161.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.034270191192627,
|
|
"epoch": 3.3600198906016905,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000427536490125486,
|
|
"loss": 5.6994,
|
|
"mean_token_accuracy": 0.17543815076351166,
|
|
"num_tokens": 6616540.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.1236707210540775,
|
|
"epoch": 3.364992541024366,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042724917553890507,
|
|
"loss": 5.8306,
|
|
"mean_token_accuracy": 0.16724002957344056,
|
|
"num_tokens": 6626109.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.184122514724732,
|
|
"epoch": 3.3699651914470414,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042696140228538616,
|
|
"loss": 5.8298,
|
|
"mean_token_accuracy": 0.16748119443655013,
|
|
"num_tokens": 6635576.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.168194246292114,
|
|
"epoch": 3.3749378418697167,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042667317123187696,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.16555749028921127,
|
|
"num_tokens": 6645728.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.176511907577515,
|
|
"epoch": 3.379910492292392,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042638448324670424,
|
|
"loss": 5.7922,
|
|
"mean_token_accuracy": 0.16968945115804673,
|
|
"num_tokens": 6654538.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.130993366241455,
|
|
"epoch": 3.384883142715067,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004260953391995716,
|
|
"loss": 5.7541,
|
|
"mean_token_accuracy": 0.17084741443395615,
|
|
"num_tokens": 6664811.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.119839000701904,
|
|
"epoch": 3.3898557931377424,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042580573996155646,
|
|
"loss": 5.8762,
|
|
"mean_token_accuracy": 0.1701633796095848,
|
|
"num_tokens": 6674185.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.206827974319458,
|
|
"epoch": 3.3948284435604177,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004255156864051074,
|
|
"loss": 5.8645,
|
|
"mean_token_accuracy": 0.1577596679329872,
|
|
"num_tokens": 6684103.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.982978487014771,
|
|
"epoch": 3.399801093983093,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042522517940404186,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.19184768497943877,
|
|
"num_tokens": 6693631.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.12087869644165,
|
|
"epoch": 3.404773744405768,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004249342198335433,
|
|
"loss": 5.8995,
|
|
"mean_token_accuracy": 0.15990070104599,
|
|
"num_tokens": 6704465.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.136934995651245,
|
|
"epoch": 3.4097463948284434,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042464280857015856,
|
|
"loss": 5.8614,
|
|
"mean_token_accuracy": 0.16950390934944154,
|
|
"num_tokens": 6713592.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.1788756370544435,
|
|
"epoch": 3.4147190452511187,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042435094649179525,
|
|
"loss": 5.8922,
|
|
"mean_token_accuracy": 0.1600611574947834,
|
|
"num_tokens": 6726252.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.197855663299561,
|
|
"epoch": 3.419691695673794,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004240586344777191,
|
|
"loss": 5.9647,
|
|
"mean_token_accuracy": 0.15676831752061843,
|
|
"num_tokens": 6736301.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.214689064025879,
|
|
"epoch": 3.4246643460964696,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004237658734085514,
|
|
"loss": 5.9164,
|
|
"mean_token_accuracy": 0.15713092535734177,
|
|
"num_tokens": 6745706.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.1950023651123045,
|
|
"epoch": 3.429636996519145,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042347266416626624,
|
|
"loss": 5.9101,
|
|
"mean_token_accuracy": 0.15741328597068788,
|
|
"num_tokens": 6755811.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.2221887588500975,
|
|
"epoch": 3.43460964694182,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004231790076341877,
|
|
"loss": 5.8487,
|
|
"mean_token_accuracy": 0.16278073266148568,
|
|
"num_tokens": 6766365.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.13021445274353,
|
|
"epoch": 3.4395822973644954,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004228849046969876,
|
|
"loss": 5.8101,
|
|
"mean_token_accuracy": 0.17333039045333862,
|
|
"num_tokens": 6775806.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.175165224075317,
|
|
"epoch": 3.4445549477871706,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004225903562406826,
|
|
"loss": 5.8553,
|
|
"mean_token_accuracy": 0.16729195415973663,
|
|
"num_tokens": 6785996.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.1441749095916744,
|
|
"epoch": 3.449527598209846,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004222953631526313,
|
|
"loss": 5.8115,
|
|
"mean_token_accuracy": 0.17142365872859955,
|
|
"num_tokens": 6795528.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.121980428695679,
|
|
"epoch": 3.454500248632521,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000421999926321532,
|
|
"loss": 5.8133,
|
|
"mean_token_accuracy": 0.16776325851678847,
|
|
"num_tokens": 6804592.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.115097618103027,
|
|
"epoch": 3.4594728990551964,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004217040466374199,
|
|
"loss": 5.8486,
|
|
"mean_token_accuracy": 0.16465378999710084,
|
|
"num_tokens": 6813242.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.193531513214111,
|
|
"epoch": 3.4644455494778716,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042140772499166403,
|
|
"loss": 5.8787,
|
|
"mean_token_accuracy": 0.16703593283891677,
|
|
"num_tokens": 6823223.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.021037626266479,
|
|
"epoch": 3.469418199900547,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004211109622769652,
|
|
"loss": 5.7212,
|
|
"mean_token_accuracy": 0.1808927208185196,
|
|
"num_tokens": 6832425.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.110166788101196,
|
|
"epoch": 3.474390850323222,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042081375938735283,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.15881891250610353,
|
|
"num_tokens": 6842549.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.163953685760498,
|
|
"epoch": 3.4793635007458974,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00042051611721818233,
|
|
"loss": 5.9806,
|
|
"mean_token_accuracy": 0.15821933150291442,
|
|
"num_tokens": 6851652.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.4793635007458974,
|
|
"eval_entropy": 6.101178498861427,
|
|
"eval_loss": 6.653561115264893,
|
|
"eval_mean_token_accuracy": 0.13842429242135468,
|
|
"eval_num_tokens": 6851652.0,
|
|
"eval_runtime": 12.5856,
|
|
"eval_samples_per_second": 275.314,
|
|
"eval_steps_per_second": 34.484,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.1321413040161135,
|
|
"epoch": 3.4843361511685726,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042021803666613273,
|
|
"loss": 5.8288,
|
|
"mean_token_accuracy": 0.1699964091181755,
|
|
"num_tokens": 6860716.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.11534161567688,
|
|
"epoch": 3.489308801591248,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041991951862920365,
|
|
"loss": 5.826,
|
|
"mean_token_accuracy": 0.16513049453496934,
|
|
"num_tokens": 6870074.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.2759411334991455,
|
|
"epoch": 3.4942814520139236,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004196205640067124,
|
|
"loss": 6.0601,
|
|
"mean_token_accuracy": 0.1471278190612793,
|
|
"num_tokens": 6880569.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.273445081710816,
|
|
"epoch": 3.499254102436599,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000419321173699292,
|
|
"loss": 5.9453,
|
|
"mean_token_accuracy": 0.15566849261522292,
|
|
"num_tokens": 6890522.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.141996908187866,
|
|
"epoch": 3.504226752859274,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004190213486088877,
|
|
"loss": 5.8758,
|
|
"mean_token_accuracy": 0.16566115766763687,
|
|
"num_tokens": 6900388.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.136771726608276,
|
|
"epoch": 3.5091994032819493,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004187210896387546,
|
|
"loss": 5.8524,
|
|
"mean_token_accuracy": 0.16573458313941955,
|
|
"num_tokens": 6910467.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.103463125228882,
|
|
"epoch": 3.5141720537046246,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004184203976934552,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.1685638040304184,
|
|
"num_tokens": 6919370.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.100522851943969,
|
|
"epoch": 3.5191447041273,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041811927367885605,
|
|
"loss": 5.6798,
|
|
"mean_token_accuracy": 0.18396053314208985,
|
|
"num_tokens": 6929688.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.150850677490235,
|
|
"epoch": 3.524117354549975,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004178177185021254,
|
|
"loss": 5.9164,
|
|
"mean_token_accuracy": 0.16440291255712508,
|
|
"num_tokens": 6939861.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.136647796630859,
|
|
"epoch": 3.5290900049726504,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004175157330717305,
|
|
"loss": 5.942,
|
|
"mean_token_accuracy": 0.15268768668174743,
|
|
"num_tokens": 6949746.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.2267578125,
|
|
"epoch": 3.5340626553953256,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000417213318297435,
|
|
"loss": 5.8801,
|
|
"mean_token_accuracy": 0.16175060197710991,
|
|
"num_tokens": 6961408.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.157064390182495,
|
|
"epoch": 3.539035305818001,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041691047509029555,
|
|
"loss": 5.9279,
|
|
"mean_token_accuracy": 0.16235635727643966,
|
|
"num_tokens": 6972003.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.1385925769805905,
|
|
"epoch": 3.5440079562406765,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041660720436265987,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.16416120976209642,
|
|
"num_tokens": 6983528.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.146167278289795,
|
|
"epoch": 3.548980606663352,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004163035070281634,
|
|
"loss": 5.8992,
|
|
"mean_token_accuracy": 0.16486466974020003,
|
|
"num_tokens": 6992766.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.147321462631226,
|
|
"epoch": 3.553953257086027,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041599938400172686,
|
|
"loss": 5.7948,
|
|
"mean_token_accuracy": 0.1655216246843338,
|
|
"num_tokens": 7001797.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.1855062484741214,
|
|
"epoch": 3.5589259075087023,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041569483619955354,
|
|
"loss": 5.8345,
|
|
"mean_token_accuracy": 0.1674937129020691,
|
|
"num_tokens": 7011215.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.104428768157959,
|
|
"epoch": 3.5638985579313776,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041538986453912616,
|
|
"loss": 5.8625,
|
|
"mean_token_accuracy": 0.15826353430747986,
|
|
"num_tokens": 7021513.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.129838180541992,
|
|
"epoch": 3.568871208354053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004150844699392045,
|
|
"loss": 5.8832,
|
|
"mean_token_accuracy": 0.16164344400167466,
|
|
"num_tokens": 7031654.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.076186752319336,
|
|
"epoch": 3.573843858776728,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041477865331982264,
|
|
"loss": 5.8586,
|
|
"mean_token_accuracy": 0.1690487802028656,
|
|
"num_tokens": 7040614.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.131633329391479,
|
|
"epoch": 3.5788165091994033,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004144724156022857,
|
|
"loss": 5.8502,
|
|
"mean_token_accuracy": 0.168062262237072,
|
|
"num_tokens": 7050877.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.262466192245483,
|
|
"epoch": 3.5837891596220786,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041416575770916763,
|
|
"loss": 6.0457,
|
|
"mean_token_accuracy": 0.1510719798505306,
|
|
"num_tokens": 7061330.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.217881393432617,
|
|
"epoch": 3.588761810044754,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041385868056430827,
|
|
"loss": 5.9369,
|
|
"mean_token_accuracy": 0.16169122755527496,
|
|
"num_tokens": 7071290.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.165552091598511,
|
|
"epoch": 3.593734460467429,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004135511850928102,
|
|
"loss": 5.8965,
|
|
"mean_token_accuracy": 0.15972486585378648,
|
|
"num_tokens": 7080112.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.15502896308899,
|
|
"epoch": 3.5987071108901043,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041324327222103655,
|
|
"loss": 6.0772,
|
|
"mean_token_accuracy": 0.14863280951976776,
|
|
"num_tokens": 7091467.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.168780183792114,
|
|
"epoch": 3.6036797613127796,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00041293494287660785,
|
|
"loss": 5.6809,
|
|
"mean_token_accuracy": 0.1779902920126915,
|
|
"num_tokens": 7101744.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.170366048812866,
|
|
"epoch": 3.608652411735455,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041262619798839914,
|
|
"loss": 5.8865,
|
|
"mean_token_accuracy": 0.1630680590867996,
|
|
"num_tokens": 7112854.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.0206170082092285,
|
|
"epoch": 3.61362506215813,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00041231703848653746,
|
|
"loss": 5.8723,
|
|
"mean_token_accuracy": 0.16947200894355774,
|
|
"num_tokens": 7121613.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.147220039367676,
|
|
"epoch": 3.6185977125808053,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041200746530239903,
|
|
"loss": 5.9303,
|
|
"mean_token_accuracy": 0.16108227670192718,
|
|
"num_tokens": 7131357.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.091606330871582,
|
|
"epoch": 3.623570363003481,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004116974793686061,
|
|
"loss": 5.7845,
|
|
"mean_token_accuracy": 0.1749992474913597,
|
|
"num_tokens": 7140200.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.123060131072998,
|
|
"epoch": 3.6285430134261563,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004113870816190244,
|
|
"loss": 5.8965,
|
|
"mean_token_accuracy": 0.16507531553506852,
|
|
"num_tokens": 7149367.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.069613933563232,
|
|
"epoch": 3.6335156638488315,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004110762729887605,
|
|
"loss": 5.861,
|
|
"mean_token_accuracy": 0.1663406878709793,
|
|
"num_tokens": 7158088.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.142796754837036,
|
|
"epoch": 3.6384883142715068,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041076505441415873,
|
|
"loss": 5.8688,
|
|
"mean_token_accuracy": 0.16718580722808837,
|
|
"num_tokens": 7168141.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.199108600616455,
|
|
"epoch": 3.643460964694182,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004104534268327982,
|
|
"loss": 5.802,
|
|
"mean_token_accuracy": 0.16823348850011827,
|
|
"num_tokens": 7177646.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.181728792190552,
|
|
"epoch": 3.6484336151168573,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004101413911834904,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.15785893201828002,
|
|
"num_tokens": 7186445.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.164184522628784,
|
|
"epoch": 3.6534062655395325,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040982894840627614,
|
|
"loss": 5.8656,
|
|
"mean_token_accuracy": 0.16852291971445083,
|
|
"num_tokens": 7195812.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.080188035964966,
|
|
"epoch": 3.6583789159622078,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004095160994424228,
|
|
"loss": 5.7505,
|
|
"mean_token_accuracy": 0.17306346595287322,
|
|
"num_tokens": 7204756.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.172871828079224,
|
|
"epoch": 3.663351566384883,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040920284523442125,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.16909418255090714,
|
|
"num_tokens": 7214708.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.145456743240357,
|
|
"epoch": 3.6683242168075583,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040888918672598337,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.16054224371910095,
|
|
"num_tokens": 7224623.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.122754859924316,
|
|
"epoch": 3.673296867230234,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000408575124862039,
|
|
"loss": 5.8588,
|
|
"mean_token_accuracy": 0.1661542534828186,
|
|
"num_tokens": 7235228.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.153800439834595,
|
|
"epoch": 3.678269517652909,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004082606605887331,
|
|
"loss": 5.9436,
|
|
"mean_token_accuracy": 0.167875836789608,
|
|
"num_tokens": 7245318.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.926960182189942,
|
|
"epoch": 3.6832421680755845,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040794579485342304,
|
|
"loss": 5.509,
|
|
"mean_token_accuracy": 0.20385562926530837,
|
|
"num_tokens": 7255468.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.149741554260254,
|
|
"epoch": 3.6882148184982597,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040763052860467547,
|
|
"loss": 5.8962,
|
|
"mean_token_accuracy": 0.16235211789608,
|
|
"num_tokens": 7265006.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.141793537139892,
|
|
"epoch": 3.693187468920935,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004073148627922637,
|
|
"loss": 5.8738,
|
|
"mean_token_accuracy": 0.16270871758460997,
|
|
"num_tokens": 7273835.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.1927296161651615,
|
|
"epoch": 3.69816011934361,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004069987983671648,
|
|
"loss": 5.9634,
|
|
"mean_token_accuracy": 0.1557069182395935,
|
|
"num_tokens": 7284617.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.012911176681518,
|
|
"epoch": 3.7031327697662855,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040668233628155664,
|
|
"loss": 5.7069,
|
|
"mean_token_accuracy": 0.18261758536100386,
|
|
"num_tokens": 7293707.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.171784734725952,
|
|
"epoch": 3.7081054201889607,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004063654774888153,
|
|
"loss": 5.9291,
|
|
"mean_token_accuracy": 0.15238013565540315,
|
|
"num_tokens": 7303227.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.051505470275879,
|
|
"epoch": 3.713078070611636,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004060482229435116,
|
|
"loss": 5.7939,
|
|
"mean_token_accuracy": 0.17458019703626632,
|
|
"num_tokens": 7312782.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.102273988723755,
|
|
"epoch": 3.7180507210343112,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004057305736014089,
|
|
"loss": 5.9178,
|
|
"mean_token_accuracy": 0.1618453308939934,
|
|
"num_tokens": 7322249.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.169122219085693,
|
|
"epoch": 3.7230233714569865,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004054125304194598,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.16983803361654282,
|
|
"num_tokens": 7331076.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.101419734954834,
|
|
"epoch": 3.7279960218796617,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040509409435580355,
|
|
"loss": 5.8425,
|
|
"mean_token_accuracy": 0.1688559651374817,
|
|
"num_tokens": 7340039.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.103044891357422,
|
|
"epoch": 3.732968672302337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004047752663697627,
|
|
"loss": 5.8399,
|
|
"mean_token_accuracy": 0.1686076320707798,
|
|
"num_tokens": 7349300.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.229220533370972,
|
|
"epoch": 3.7379413227250122,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004044560474218409,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.15839385688304902,
|
|
"num_tokens": 7358363.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.103728914260865,
|
|
"epoch": 3.7429139731476875,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040413643847371923,
|
|
"loss": 5.8539,
|
|
"mean_token_accuracy": 0.17160698473453523,
|
|
"num_tokens": 7369391.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.150986289978027,
|
|
"epoch": 3.7478866235703627,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004038164404882539,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.15380954295396804,
|
|
"num_tokens": 7378272.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.110385513305664,
|
|
"epoch": 3.7528592739930384,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004034960544294731,
|
|
"loss": 5.7825,
|
|
"mean_token_accuracy": 0.172193244099617,
|
|
"num_tokens": 7387590.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.1140649795532225,
|
|
"epoch": 3.7578319244157137,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040317528126257413,
|
|
"loss": 5.8886,
|
|
"mean_token_accuracy": 0.15803974568843843,
|
|
"num_tokens": 7397733.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.104284286499023,
|
|
"epoch": 3.762804574838389,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004028541219539205,
|
|
"loss": 5.7609,
|
|
"mean_token_accuracy": 0.17075671553611754,
|
|
"num_tokens": 7406748.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.118281888961792,
|
|
"epoch": 3.767777225261064,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004025325774710391,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.1596181273460388,
|
|
"num_tokens": 7416419.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.106326961517334,
|
|
"epoch": 3.7727498756837394,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040221064878261684,
|
|
"loss": 5.8184,
|
|
"mean_token_accuracy": 0.16891866624355317,
|
|
"num_tokens": 7425257.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.2067852973937985,
|
|
"epoch": 3.7777225261064147,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004018883368584985,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.16109876930713654,
|
|
"num_tokens": 7436914.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.14767599105835,
|
|
"epoch": 3.78269517652909,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040156564266968336,
|
|
"loss": 5.9226,
|
|
"mean_token_accuracy": 0.16273024380207063,
|
|
"num_tokens": 7447568.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.0983898639678955,
|
|
"epoch": 3.787667826951765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040124256718832214,
|
|
"loss": 5.8824,
|
|
"mean_token_accuracy": 0.16562072038650513,
|
|
"num_tokens": 7457717.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.164183282852173,
|
|
"epoch": 3.7926404773744404,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004009191113877141,
|
|
"loss": 5.8723,
|
|
"mean_token_accuracy": 0.16255360394716262,
|
|
"num_tokens": 7467123.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.086970329284668,
|
|
"epoch": 3.797613127797116,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004005952762423048,
|
|
"loss": 5.8411,
|
|
"mean_token_accuracy": 0.16852075457572938,
|
|
"num_tokens": 7476987.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.179725694656372,
|
|
"epoch": 3.8025857782197914,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004002710627276821,
|
|
"loss": 5.9467,
|
|
"mean_token_accuracy": 0.1670974776148796,
|
|
"num_tokens": 7487024.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.171169710159302,
|
|
"epoch": 3.8075584286424666,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003999464718205739,
|
|
"loss": 5.955,
|
|
"mean_token_accuracy": 0.15775143429636956,
|
|
"num_tokens": 7496826.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.12643780708313,
|
|
"epoch": 3.812531079065142,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003996215044988451,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.1593120887875557,
|
|
"num_tokens": 7506981.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.0427264213562015,
|
|
"epoch": 3.817503729487817,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00039929616174149454,
|
|
"loss": 5.7095,
|
|
"mean_token_accuracy": 0.17419576942920684,
|
|
"num_tokens": 7516925.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.1607279777526855,
|
|
"epoch": 3.8224763799104924,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039897044452865207,
|
|
"loss": 5.8564,
|
|
"mean_token_accuracy": 0.16927637457847594,
|
|
"num_tokens": 7525789.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.02476134300232,
|
|
"epoch": 3.8274490303331676,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003986443538415756,
|
|
"loss": 5.8001,
|
|
"mean_token_accuracy": 0.16755083799362183,
|
|
"num_tokens": 7535907.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.168830299377442,
|
|
"epoch": 3.832421680755843,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003983178906626484,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.17218476682901382,
|
|
"num_tokens": 7546985.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.021071147918701,
|
|
"epoch": 3.837394331178518,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039799105597537564,
|
|
"loss": 5.6597,
|
|
"mean_token_accuracy": 0.18736549764871596,
|
|
"num_tokens": 7556912.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.107205390930176,
|
|
"epoch": 3.8423669816011934,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003976638507643818,
|
|
"loss": 5.8845,
|
|
"mean_token_accuracy": 0.16694090366363526,
|
|
"num_tokens": 7567156.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.0821067810058596,
|
|
"epoch": 3.8473396320238686,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003973362760154076,
|
|
"loss": 5.8383,
|
|
"mean_token_accuracy": 0.1718806207180023,
|
|
"num_tokens": 7576423.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.149602174758911,
|
|
"epoch": 3.852312282446544,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003970083327153071,
|
|
"loss": 5.8297,
|
|
"mean_token_accuracy": 0.16807498633861542,
|
|
"num_tokens": 7585612.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.0555675506591795,
|
|
"epoch": 3.857284932869219,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003966800218520445,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.1683152750134468,
|
|
"num_tokens": 7595581.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.162358236312866,
|
|
"epoch": 3.8622575832918944,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003963513444146914,
|
|
"loss": 5.8432,
|
|
"mean_token_accuracy": 0.1674938380718231,
|
|
"num_tokens": 7605202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.178526496887207,
|
|
"epoch": 3.8672302337145696,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003960223013934239,
|
|
"loss": 5.7473,
|
|
"mean_token_accuracy": 0.17868545055389404,
|
|
"num_tokens": 7615783.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.18940052986145,
|
|
"epoch": 3.872202884137245,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003956928937795192,
|
|
"loss": 5.9661,
|
|
"mean_token_accuracy": 0.1520422413945198,
|
|
"num_tokens": 7626609.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.103344869613648,
|
|
"epoch": 3.8771755345599206,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000395363122565353,
|
|
"loss": 5.85,
|
|
"mean_token_accuracy": 0.16939883977174758,
|
|
"num_tokens": 7636676.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.184452629089355,
|
|
"epoch": 3.882148184982596,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039503298874439655,
|
|
"loss": 5.8461,
|
|
"mean_token_accuracy": 0.16941933780908586,
|
|
"num_tokens": 7646523.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.045930194854736,
|
|
"epoch": 3.887120835405271,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003947024933112131,
|
|
"loss": 5.7414,
|
|
"mean_token_accuracy": 0.17759622633457184,
|
|
"num_tokens": 7656267.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.103095436096192,
|
|
"epoch": 3.8920934858279463,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003943716372614556,
|
|
"loss": 5.8154,
|
|
"mean_token_accuracy": 0.16899876296520233,
|
|
"num_tokens": 7665529.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.141912317276001,
|
|
"epoch": 3.8970661362506216,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003940404215918634,
|
|
"loss": 5.8922,
|
|
"mean_token_accuracy": 0.1622321769595146,
|
|
"num_tokens": 7675477.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.910722112655639,
|
|
"epoch": 3.902038786673297,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00039370884730025897,
|
|
"loss": 5.529,
|
|
"mean_token_accuracy": 0.1907438039779663,
|
|
"num_tokens": 7685675.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.050700092315674,
|
|
"epoch": 3.907011437095972,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003933769153855455,
|
|
"loss": 5.8787,
|
|
"mean_token_accuracy": 0.17111240923404694,
|
|
"num_tokens": 7695821.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.09744930267334,
|
|
"epoch": 3.9119840875186473,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003930446268477034,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.16596696376800538,
|
|
"num_tokens": 7706163.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.098593521118164,
|
|
"epoch": 3.9169567379413226,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003927119826877875,
|
|
"loss": 5.7996,
|
|
"mean_token_accuracy": 0.17113927900791168,
|
|
"num_tokens": 7715125.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.082453870773316,
|
|
"epoch": 3.921929388363998,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003923789839079238,
|
|
"loss": 5.8044,
|
|
"mean_token_accuracy": 0.17066917717456817,
|
|
"num_tokens": 7724546.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.942547750473023,
|
|
"epoch": 3.9269020387866735,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00039204563151130683,
|
|
"loss": 5.6502,
|
|
"mean_token_accuracy": 0.18459851443767547,
|
|
"num_tokens": 7734829.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.080030488967895,
|
|
"epoch": 3.931874689209349,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003917119265021965,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.1653016597032547,
|
|
"num_tokens": 7744612.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.159715223312378,
|
|
"epoch": 3.936847339632024,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003913778698859146,
|
|
"loss": 5.9041,
|
|
"mean_token_accuracy": 0.16572495996952058,
|
|
"num_tokens": 7754042.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.146387672424316,
|
|
"epoch": 3.9418199900546993,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003910434626688427,
|
|
"loss": 5.8099,
|
|
"mean_token_accuracy": 0.1672035574913025,
|
|
"num_tokens": 7762608.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.050923824310303,
|
|
"epoch": 3.9467926404773745,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003907087058584183,
|
|
"loss": 5.7381,
|
|
"mean_token_accuracy": 0.17555425316095352,
|
|
"num_tokens": 7771477.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.067458248138427,
|
|
"epoch": 3.95176529090005,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00039037360046313186,
|
|
"loss": 5.8277,
|
|
"mean_token_accuracy": 0.17463671043515205,
|
|
"num_tokens": 7782785.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.253055238723755,
|
|
"epoch": 3.956737941322725,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039003814749252454,
|
|
"loss": 6.0167,
|
|
"mean_token_accuracy": 0.15535970404744148,
|
|
"num_tokens": 7793572.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.102360677719116,
|
|
"epoch": 3.9617105917454003,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038970234795718424,
|
|
"loss": 5.7693,
|
|
"mean_token_accuracy": 0.16971321105957032,
|
|
"num_tokens": 7803170.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.100835180282592,
|
|
"epoch": 3.9666832421680756,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000389366202868743,
|
|
"loss": 5.9473,
|
|
"mean_token_accuracy": 0.1609964981675148,
|
|
"num_tokens": 7813445.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.1510998725891115,
|
|
"epoch": 3.971655892590751,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003890297132398739,
|
|
"loss": 5.9387,
|
|
"mean_token_accuracy": 0.16301140189170837,
|
|
"num_tokens": 7823089.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.09289698600769,
|
|
"epoch": 3.976628543013426,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003886928800842879,
|
|
"loss": 5.7641,
|
|
"mean_token_accuracy": 0.16849076002836227,
|
|
"num_tokens": 7832487.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.976628543013426,
|
|
"eval_entropy": 5.982148299019458,
|
|
"eval_loss": 6.588039875030518,
|
|
"eval_mean_token_accuracy": 0.13949383075095817,
|
|
"eval_num_tokens": 7832487.0,
|
|
"eval_runtime": 10.1652,
|
|
"eval_samples_per_second": 340.87,
|
|
"eval_steps_per_second": 42.695,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 6.0299553871154785,
|
|
"epoch": 3.9816011934361013,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038835570441673095,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.16857545971870422,
|
|
"num_tokens": 7842540.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 6.147599840164185,
|
|
"epoch": 3.9865738438587766,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003880181872529809,
|
|
"loss": 5.9173,
|
|
"mean_token_accuracy": 0.16664879769086838,
|
|
"num_tokens": 7852384.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 6.10081524848938,
|
|
"epoch": 3.991546494281452,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003876803296098443,
|
|
"loss": 5.8943,
|
|
"mean_token_accuracy": 0.1691052332520485,
|
|
"num_tokens": 7864082.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.133717393875122,
|
|
"epoch": 3.996519144704127,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003873421325051535,
|
|
"loss": 5.8102,
|
|
"mean_token_accuracy": 0.1647699512541294,
|
|
"num_tokens": 7873677.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 6.130531681908502,
|
|
"epoch": 4.000994530084535,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038700359695776346,
|
|
"loss": 5.7928,
|
|
"mean_token_accuracy": 0.16694999072286817,
|
|
"num_tokens": 7882368.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.062766361236572,
|
|
"epoch": 4.00596718050721,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003866647239875487,
|
|
"loss": 5.4637,
|
|
"mean_token_accuracy": 0.18448322415351867,
|
|
"num_tokens": 7892327.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 6.041069507598877,
|
|
"epoch": 4.010939830929885,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038632551461540023,
|
|
"loss": 5.4634,
|
|
"mean_token_accuracy": 0.1840252920985222,
|
|
"num_tokens": 7902048.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 6.111983919143677,
|
|
"epoch": 4.015912481352561,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003859859698632228,
|
|
"loss": 5.5086,
|
|
"mean_token_accuracy": 0.1838537022471428,
|
|
"num_tokens": 7911512.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.08398585319519,
|
|
"epoch": 4.020885131775236,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003856460907539312,
|
|
"loss": 5.5992,
|
|
"mean_token_accuracy": 0.1730125829577446,
|
|
"num_tokens": 7921832.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.077848100662232,
|
|
"epoch": 4.025857782197911,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003853058783114478,
|
|
"loss": 5.5052,
|
|
"mean_token_accuracy": 0.18308013081550598,
|
|
"num_tokens": 7931143.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 6.007107877731324,
|
|
"epoch": 4.030830432620586,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003849653335606987,
|
|
"loss": 5.4098,
|
|
"mean_token_accuracy": 0.19528655111789703,
|
|
"num_tokens": 7941108.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.991386938095093,
|
|
"epoch": 4.035803083043262,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003846244575276117,
|
|
"loss": 5.4768,
|
|
"mean_token_accuracy": 0.18108234852552413,
|
|
"num_tokens": 7951464.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.174937534332275,
|
|
"epoch": 4.040775733465938,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038428325123911205,
|
|
"loss": 5.5954,
|
|
"mean_token_accuracy": 0.17571333348751067,
|
|
"num_tokens": 7961626.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.089084959030151,
|
|
"epoch": 4.045748383888613,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003839417157231205,
|
|
"loss": 5.5806,
|
|
"mean_token_accuracy": 0.17940284162759781,
|
|
"num_tokens": 7970135.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.965398454666138,
|
|
"epoch": 4.050721034311288,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003835998520085492,
|
|
"loss": 5.5072,
|
|
"mean_token_accuracy": 0.18380907773971558,
|
|
"num_tokens": 7982177.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 6.059826278686524,
|
|
"epoch": 4.0556936847339635,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003832576611252993,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.18728707134723663,
|
|
"num_tokens": 7991845.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 6.021476316452026,
|
|
"epoch": 4.060666335156639,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003829151441042572,
|
|
"loss": 5.4403,
|
|
"mean_token_accuracy": 0.18769097328186035,
|
|
"num_tokens": 8000731.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 6.038483381271362,
|
|
"epoch": 4.065638985579314,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038257230197729237,
|
|
"loss": 5.4739,
|
|
"mean_token_accuracy": 0.18641373813152312,
|
|
"num_tokens": 8011137.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 6.046646928787231,
|
|
"epoch": 4.070611636001989,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038222913577725325,
|
|
"loss": 5.5234,
|
|
"mean_token_accuracy": 0.1738094136118889,
|
|
"num_tokens": 8020413.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 6.011880016326904,
|
|
"epoch": 4.0755842864246645,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038188564653796487,
|
|
"loss": 5.4632,
|
|
"mean_token_accuracy": 0.19438748359680175,
|
|
"num_tokens": 8029727.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 6.042710828781128,
|
|
"epoch": 4.08055693684734,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003815418352942252,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.1855413183569908,
|
|
"num_tokens": 8038707.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.885494375228882,
|
|
"epoch": 4.085529587270015,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003811977030818026,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.21039735972881318,
|
|
"num_tokens": 8048604.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.99918041229248,
|
|
"epoch": 4.09050223769269,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000380853250937432,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.18792495429515838,
|
|
"num_tokens": 8058089.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 6.151983118057251,
|
|
"epoch": 4.0954748881153655,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038050847989881236,
|
|
"loss": 5.5182,
|
|
"mean_token_accuracy": 0.18131167441606522,
|
|
"num_tokens": 8068078.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.955233764648438,
|
|
"epoch": 4.100447538538041,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003801633910046035,
|
|
"loss": 5.3959,
|
|
"mean_token_accuracy": 0.19530741274356841,
|
|
"num_tokens": 8076918.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.864402389526367,
|
|
"epoch": 4.105420188960716,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003798179852944225,
|
|
"loss": 5.3072,
|
|
"mean_token_accuracy": 0.20215623080730438,
|
|
"num_tokens": 8087104.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.994411659240723,
|
|
"epoch": 4.110392839383391,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037947226380884115,
|
|
"loss": 5.5833,
|
|
"mean_token_accuracy": 0.17266546040773392,
|
|
"num_tokens": 8096976.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 6.040523052215576,
|
|
"epoch": 4.1153654898060665,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037912622758938243,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.18574922680854797,
|
|
"num_tokens": 8106707.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 6.049013757705689,
|
|
"epoch": 4.120338140228742,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003787798776785174,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.17531205266714095,
|
|
"num_tokens": 8116822.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.907098197937012,
|
|
"epoch": 4.125310790651417,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003784332151196622,
|
|
"loss": 5.4359,
|
|
"mean_token_accuracy": 0.1902480751276016,
|
|
"num_tokens": 8125677.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.837154293060303,
|
|
"epoch": 4.130283441074092,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037808624095717503,
|
|
"loss": 5.3635,
|
|
"mean_token_accuracy": 0.19794866293668748,
|
|
"num_tokens": 8134878.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 6.002427005767823,
|
|
"epoch": 4.1352560914967675,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003777389562363526,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.19128775596618652,
|
|
"num_tokens": 8144491.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 6.024952554702759,
|
|
"epoch": 4.140228741919443,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037739136200342743,
|
|
"loss": 5.5422,
|
|
"mean_token_accuracy": 0.17141647338867189,
|
|
"num_tokens": 8155163.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 6.011817836761475,
|
|
"epoch": 4.145201392342118,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037704345930556426,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.176428160071373,
|
|
"num_tokens": 8164626.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.105657434463501,
|
|
"epoch": 4.150174042764793,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037669524919085716,
|
|
"loss": 5.6443,
|
|
"mean_token_accuracy": 0.17440564185380936,
|
|
"num_tokens": 8174681.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.12515869140625,
|
|
"epoch": 4.1551466931874685,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003763467327083264,
|
|
"loss": 5.7017,
|
|
"mean_token_accuracy": 0.17376572489738465,
|
|
"num_tokens": 8184627.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.943635082244873,
|
|
"epoch": 4.160119343610144,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037599791090791527,
|
|
"loss": 5.3436,
|
|
"mean_token_accuracy": 0.1976415768265724,
|
|
"num_tokens": 8195029.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.976450204849243,
|
|
"epoch": 4.165091994032819,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.0003756487848404868,
|
|
"loss": 5.3918,
|
|
"mean_token_accuracy": 0.20074576884508133,
|
|
"num_tokens": 8204897.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 6.002750492095947,
|
|
"epoch": 4.170064644455495,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003752993555578205,
|
|
"loss": 5.6282,
|
|
"mean_token_accuracy": 0.17548299282789231,
|
|
"num_tokens": 8215859.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.961316823959351,
|
|
"epoch": 4.17503729487817,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003749496241126095,
|
|
"loss": 5.4177,
|
|
"mean_token_accuracy": 0.1890671357512474,
|
|
"num_tokens": 8225377.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 6.016830730438232,
|
|
"epoch": 4.180009945300846,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037459959155845726,
|
|
"loss": 5.6188,
|
|
"mean_token_accuracy": 0.17079947143793106,
|
|
"num_tokens": 8235436.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 6.030125522613526,
|
|
"epoch": 4.184982595723521,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003742492589498744,
|
|
"loss": 5.5349,
|
|
"mean_token_accuracy": 0.18335505127906798,
|
|
"num_tokens": 8245129.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.976276874542236,
|
|
"epoch": 4.189955246146196,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037389862734227525,
|
|
"loss": 5.4833,
|
|
"mean_token_accuracy": 0.18531831204891205,
|
|
"num_tokens": 8254870.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 6.053520441055298,
|
|
"epoch": 4.194927896568871,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037354769779197514,
|
|
"loss": 5.5655,
|
|
"mean_token_accuracy": 0.18109067976474763,
|
|
"num_tokens": 8266078.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.09083776473999,
|
|
"epoch": 4.199900546991547,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037319647135618695,
|
|
"loss": 5.5858,
|
|
"mean_token_accuracy": 0.18003323525190354,
|
|
"num_tokens": 8275833.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.922969102859497,
|
|
"epoch": 4.204873197414222,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037284494909301793,
|
|
"loss": 5.487,
|
|
"mean_token_accuracy": 0.1761337324976921,
|
|
"num_tokens": 8286866.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 6.0536754608154295,
|
|
"epoch": 4.209845847836897,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003724931320614663,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.18408893048763275,
|
|
"num_tokens": 8296316.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 6.1041261672973635,
|
|
"epoch": 4.214818498259572,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037214102132141874,
|
|
"loss": 5.6726,
|
|
"mean_token_accuracy": 0.17136690244078637,
|
|
"num_tokens": 8306023.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.894186925888062,
|
|
"epoch": 4.219791148682248,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003717886179336465,
|
|
"loss": 5.4267,
|
|
"mean_token_accuracy": 0.18311179131269456,
|
|
"num_tokens": 8315808.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 6.018088150024414,
|
|
"epoch": 4.224763799104923,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037143592295980243,
|
|
"loss": 5.5135,
|
|
"mean_token_accuracy": 0.1837345242500305,
|
|
"num_tokens": 8324951.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.968040895462036,
|
|
"epoch": 4.229736449527598,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003710829374624181,
|
|
"loss": 5.5109,
|
|
"mean_token_accuracy": 0.1846233308315277,
|
|
"num_tokens": 8335019.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 6.053265047073364,
|
|
"epoch": 4.2347090999502734,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003707296625048999,
|
|
"loss": 5.487,
|
|
"mean_token_accuracy": 0.18402136266231536,
|
|
"num_tokens": 8344496.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 6.039746856689453,
|
|
"epoch": 4.239681750372949,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003703760991515264,
|
|
"loss": 5.5338,
|
|
"mean_token_accuracy": 0.17707462310791017,
|
|
"num_tokens": 8353766.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 6.03184700012207,
|
|
"epoch": 4.244654400795624,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037002224846744526,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.1743014246225357,
|
|
"num_tokens": 8363576.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.124289846420288,
|
|
"epoch": 4.249627051218299,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036966811151866943,
|
|
"loss": 5.6041,
|
|
"mean_token_accuracy": 0.17384740114212036,
|
|
"num_tokens": 8372714.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.911302089691162,
|
|
"epoch": 4.2545997016409745,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036931368937207436,
|
|
"loss": 5.5126,
|
|
"mean_token_accuracy": 0.18899890929460525,
|
|
"num_tokens": 8383563.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.972736072540283,
|
|
"epoch": 4.25957235206365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003689589830953948,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.19609167724847792,
|
|
"num_tokens": 8393352.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 6.010274219512939,
|
|
"epoch": 4.264545002486325,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003686039937572213,
|
|
"loss": 5.5725,
|
|
"mean_token_accuracy": 0.17558750063180922,
|
|
"num_tokens": 8403219.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.932902145385742,
|
|
"epoch": 4.269517652909,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003682487224269972,
|
|
"loss": 5.421,
|
|
"mean_token_accuracy": 0.19294578433036805,
|
|
"num_tokens": 8411596.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.966885280609131,
|
|
"epoch": 4.2744903033316755,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003678931701750155,
|
|
"loss": 5.5301,
|
|
"mean_token_accuracy": 0.18093415349721909,
|
|
"num_tokens": 8421472.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.987767314910888,
|
|
"epoch": 4.279462953754351,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00036753733807241536,
|
|
"loss": 5.4486,
|
|
"mean_token_accuracy": 0.19371780902147293,
|
|
"num_tokens": 8431489.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.804828214645386,
|
|
"epoch": 4.284435604177026,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00036718122719117907,
|
|
"loss": 5.2864,
|
|
"mean_token_accuracy": 0.20603837966918945,
|
|
"num_tokens": 8441108.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.866740942001343,
|
|
"epoch": 4.289408254599701,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003668248386041288,
|
|
"loss": 5.3548,
|
|
"mean_token_accuracy": 0.19954095482826234,
|
|
"num_tokens": 8451664.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 6.007674837112427,
|
|
"epoch": 4.294380905022377,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036646817338492334,
|
|
"loss": 5.5336,
|
|
"mean_token_accuracy": 0.18146137297153472,
|
|
"num_tokens": 8460531.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 6.013134717941284,
|
|
"epoch": 4.299353555445053,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003661112326080548,
|
|
"loss": 5.5111,
|
|
"mean_token_accuracy": 0.1885731890797615,
|
|
"num_tokens": 8470018.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.92304310798645,
|
|
"epoch": 4.304326205867728,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003657540173488455,
|
|
"loss": 5.4443,
|
|
"mean_token_accuracy": 0.19588665366172792,
|
|
"num_tokens": 8480095.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 6.008934688568115,
|
|
"epoch": 4.309298856290403,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003653965286834446,
|
|
"loss": 5.6116,
|
|
"mean_token_accuracy": 0.17072625309228898,
|
|
"num_tokens": 8489817.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 6.097415590286255,
|
|
"epoch": 4.314271506713078,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003650387676888251,
|
|
"loss": 5.6069,
|
|
"mean_token_accuracy": 0.17216630280017853,
|
|
"num_tokens": 8499333.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 6.059564781188965,
|
|
"epoch": 4.319244157135754,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003646807354427802,
|
|
"loss": 5.5823,
|
|
"mean_token_accuracy": 0.1695585936307907,
|
|
"num_tokens": 8508732.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.9678082942962645,
|
|
"epoch": 4.324216807558429,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036432243302392025,
|
|
"loss": 5.5687,
|
|
"mean_token_accuracy": 0.17253459244966507,
|
|
"num_tokens": 8519525.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.914106607437134,
|
|
"epoch": 4.329189457981104,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00036396386151166975,
|
|
"loss": 5.443,
|
|
"mean_token_accuracy": 0.19504293501377107,
|
|
"num_tokens": 8529559.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 6.008754253387451,
|
|
"epoch": 4.334162108403779,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003636050219862638,
|
|
"loss": 5.4972,
|
|
"mean_token_accuracy": 0.17805957049131393,
|
|
"num_tokens": 8538615.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.957222938537598,
|
|
"epoch": 4.339134758826455,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036324591552874493,
|
|
"loss": 5.4611,
|
|
"mean_token_accuracy": 0.18662677109241485,
|
|
"num_tokens": 8548432.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.924053716659546,
|
|
"epoch": 4.34410740924913,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036288654322095973,
|
|
"loss": 5.434,
|
|
"mean_token_accuracy": 0.19401901066303254,
|
|
"num_tokens": 8558216.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.962054252624512,
|
|
"epoch": 4.349080059671805,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036252690614555557,
|
|
"loss": 5.3805,
|
|
"mean_token_accuracy": 0.18866126388311386,
|
|
"num_tokens": 8568234.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.962238550186157,
|
|
"epoch": 4.35405271009448,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003621670053859778,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.1772398978471756,
|
|
"num_tokens": 8578140.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 6.036499977111816,
|
|
"epoch": 4.359025360517156,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000361806842026466,
|
|
"loss": 5.5548,
|
|
"mean_token_accuracy": 0.1814563199877739,
|
|
"num_tokens": 8587976.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.96282639503479,
|
|
"epoch": 4.363998010939831,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00036144641715205077,
|
|
"loss": 5.5553,
|
|
"mean_token_accuracy": 0.18535214215517043,
|
|
"num_tokens": 8597739.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.994455623626709,
|
|
"epoch": 4.368970661362506,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003610857318485507,
|
|
"loss": 5.6083,
|
|
"mean_token_accuracy": 0.17998283505439758,
|
|
"num_tokens": 8608609.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.983720302581787,
|
|
"epoch": 4.373943311785181,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036072478720256885,
|
|
"loss": 5.54,
|
|
"mean_token_accuracy": 0.17714957594871522,
|
|
"num_tokens": 8617870.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.984472751617432,
|
|
"epoch": 4.378915962207857,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003603635843014895,
|
|
"loss": 5.5041,
|
|
"mean_token_accuracy": 0.18702592998743056,
|
|
"num_tokens": 8627811.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 6.034168720245361,
|
|
"epoch": 4.383888612630532,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036000212423347523,
|
|
"loss": 5.5934,
|
|
"mean_token_accuracy": 0.1801157549023628,
|
|
"num_tokens": 8636989.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.9784081935882565,
|
|
"epoch": 4.388861263053207,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000359640408087463,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.1897987276315689,
|
|
"num_tokens": 8646090.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.984687852859497,
|
|
"epoch": 4.393833913475882,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003592784369531615,
|
|
"loss": 5.4458,
|
|
"mean_token_accuracy": 0.18166970014572142,
|
|
"num_tokens": 8656244.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.836618185043335,
|
|
"epoch": 4.398806563898558,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035891621192104764,
|
|
"loss": 5.4212,
|
|
"mean_token_accuracy": 0.19568043649196626,
|
|
"num_tokens": 8665783.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 6.025731611251831,
|
|
"epoch": 4.403779214321233,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035855373408236294,
|
|
"loss": 5.6127,
|
|
"mean_token_accuracy": 0.1750440776348114,
|
|
"num_tokens": 8676736.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 6.031400299072265,
|
|
"epoch": 4.408751864743908,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003581910045291107,
|
|
"loss": 5.5621,
|
|
"mean_token_accuracy": 0.18004930764436722,
|
|
"num_tokens": 8687035.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 6.034171104431152,
|
|
"epoch": 4.413724515166583,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003578280243540526,
|
|
"loss": 5.5881,
|
|
"mean_token_accuracy": 0.18103364408016204,
|
|
"num_tokens": 8696939.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.958732986450196,
|
|
"epoch": 4.4186971655892595,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00035746479465070515,
|
|
"loss": 5.5694,
|
|
"mean_token_accuracy": 0.18149563819169998,
|
|
"num_tokens": 8707213.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.975217342376709,
|
|
"epoch": 4.423669816011935,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00035710131651333683,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.18179885149002076,
|
|
"num_tokens": 8717066.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 6.013613557815551,
|
|
"epoch": 4.42864246643461,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035673759103696447,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.18129101246595383,
|
|
"num_tokens": 8727421.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 6.019707202911377,
|
|
"epoch": 4.433615116857285,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003563736193173499,
|
|
"loss": 5.5704,
|
|
"mean_token_accuracy": 0.17601671069860458,
|
|
"num_tokens": 8736086.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 6.093293762207031,
|
|
"epoch": 4.4385877672799605,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003560094024509969,
|
|
"loss": 5.6113,
|
|
"mean_token_accuracy": 0.18244358003139496,
|
|
"num_tokens": 8746496.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.9355121612548825,
|
|
"epoch": 4.443560417702636,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00035564494153514777,
|
|
"loss": 5.4661,
|
|
"mean_token_accuracy": 0.19450794905424118,
|
|
"num_tokens": 8755528.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.993637132644653,
|
|
"epoch": 4.448533068125311,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035528023766778,
|
|
"loss": 5.5591,
|
|
"mean_token_accuracy": 0.17501043826341628,
|
|
"num_tokens": 8765486.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 6.039368057250977,
|
|
"epoch": 4.453505718547986,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003549152919476031,
|
|
"loss": 5.5855,
|
|
"mean_token_accuracy": 0.177178293466568,
|
|
"num_tokens": 8775315.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 6.006639242172241,
|
|
"epoch": 4.4584783689706615,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035455010547405513,
|
|
"loss": 5.592,
|
|
"mean_token_accuracy": 0.17521093487739564,
|
|
"num_tokens": 8784976.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 6.003523969650269,
|
|
"epoch": 4.463451019393337,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035418467934729934,
|
|
"loss": 5.6192,
|
|
"mean_token_accuracy": 0.17238331735134124,
|
|
"num_tokens": 8795412.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.951861095428467,
|
|
"epoch": 4.468423669816012,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00035381901466822113,
|
|
"loss": 5.4897,
|
|
"mean_token_accuracy": 0.1896444723010063,
|
|
"num_tokens": 8805277.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.965359163284302,
|
|
"epoch": 4.473396320238687,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003534531125384244,
|
|
"loss": 5.5337,
|
|
"mean_token_accuracy": 0.18066484928131105,
|
|
"num_tokens": 8814308.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.473396320238687,
|
|
"eval_entropy": 5.889368010006742,
|
|
"eval_loss": 6.584246635437012,
|
|
"eval_mean_token_accuracy": 0.13975645711619733,
|
|
"eval_num_tokens": 8814308.0,
|
|
"eval_runtime": 11.4052,
|
|
"eval_samples_per_second": 303.808,
|
|
"eval_steps_per_second": 38.053,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.965989971160889,
|
|
"epoch": 4.4783689706613625,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035308697406022865,
|
|
"loss": 5.5889,
|
|
"mean_token_accuracy": 0.1797664001584053,
|
|
"num_tokens": 8824836.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.876661682128907,
|
|
"epoch": 4.483341621084038,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035272060033666503,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.18881964683532715,
|
|
"num_tokens": 8834390.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 6.081400966644287,
|
|
"epoch": 4.488314271506713,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003523539924714738,
|
|
"loss": 5.6218,
|
|
"mean_token_accuracy": 0.1754921093583107,
|
|
"num_tokens": 8844266.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 6.031957483291626,
|
|
"epoch": 4.493286921929388,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035198715156910016,
|
|
"loss": 5.5207,
|
|
"mean_token_accuracy": 0.18328167796134948,
|
|
"num_tokens": 8852876.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.759106874465942,
|
|
"epoch": 4.4982595723520635,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035162007873469175,
|
|
"loss": 5.224,
|
|
"mean_token_accuracy": 0.21215859055519104,
|
|
"num_tokens": 8862861.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.939958000183106,
|
|
"epoch": 4.503232222774739,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035125277507409483,
|
|
"loss": 5.5039,
|
|
"mean_token_accuracy": 0.1848140150308609,
|
|
"num_tokens": 8872177.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.884758234024048,
|
|
"epoch": 4.508204873197414,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00035088524169385075,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.18632982224225997,
|
|
"num_tokens": 8882202.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.98185019493103,
|
|
"epoch": 4.513177523620089,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00035051747970119354,
|
|
"loss": 5.538,
|
|
"mean_token_accuracy": 0.18283590525388718,
|
|
"num_tokens": 8892361.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 6.04368839263916,
|
|
"epoch": 4.5181501740427645,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035014949020404523,
|
|
"loss": 5.6101,
|
|
"mean_token_accuracy": 0.17518611997365952,
|
|
"num_tokens": 8903014.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.995471954345703,
|
|
"epoch": 4.52312282446544,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034978127431101365,
|
|
"loss": 5.5949,
|
|
"mean_token_accuracy": 0.17505080699920655,
|
|
"num_tokens": 8912145.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.96468448638916,
|
|
"epoch": 4.528095474888115,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034941283313138865,
|
|
"loss": 5.519,
|
|
"mean_token_accuracy": 0.18279676437377929,
|
|
"num_tokens": 8921748.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 6.014121103286743,
|
|
"epoch": 4.53306812531079,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003490441677751387,
|
|
"loss": 5.5943,
|
|
"mean_token_accuracy": 0.1770382985472679,
|
|
"num_tokens": 8932327.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.968025064468383,
|
|
"epoch": 4.5380407757334655,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00034867527935290773,
|
|
"loss": 5.6315,
|
|
"mean_token_accuracy": 0.1771518349647522,
|
|
"num_tokens": 8942781.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 6.044249677658081,
|
|
"epoch": 4.543013426156142,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00034830616897601146,
|
|
"loss": 5.5698,
|
|
"mean_token_accuracy": 0.17739802300930024,
|
|
"num_tokens": 8951887.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 6.051941967010498,
|
|
"epoch": 4.547986076578816,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003479368377564344,
|
|
"loss": 5.6207,
|
|
"mean_token_accuracy": 0.1737745612859726,
|
|
"num_tokens": 8961454.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.9568479537963865,
|
|
"epoch": 4.552958727001492,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003475672868068265,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.18440057337284088,
|
|
"num_tokens": 8970218.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.985123491287231,
|
|
"epoch": 4.557931377424167,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034719751724049955,
|
|
"loss": 5.5115,
|
|
"mean_token_accuracy": 0.18447001427412033,
|
|
"num_tokens": 8980141.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.954796648025512,
|
|
"epoch": 4.562904027846843,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00034682753017142387,
|
|
"loss": 5.5144,
|
|
"mean_token_accuracy": 0.18636007905006408,
|
|
"num_tokens": 8989514.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.972196626663208,
|
|
"epoch": 4.567876678269518,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034645732671422517,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.1778627559542656,
|
|
"num_tokens": 9000054.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 6.0173228740692135,
|
|
"epoch": 4.572849328692193,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003460869079841809,
|
|
"loss": 5.6219,
|
|
"mean_token_accuracy": 0.18261063247919082,
|
|
"num_tokens": 9010070.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 6.0533064842224125,
|
|
"epoch": 4.577821979114868,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003457162750972172,
|
|
"loss": 5.6019,
|
|
"mean_token_accuracy": 0.17537229955196382,
|
|
"num_tokens": 9019785.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.979168128967285,
|
|
"epoch": 4.582794629537544,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034534542916990534,
|
|
"loss": 5.5988,
|
|
"mean_token_accuracy": 0.1796376332640648,
|
|
"num_tokens": 9028281.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.847357749938965,
|
|
"epoch": 4.587767279960219,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034497437131945837,
|
|
"loss": 5.3574,
|
|
"mean_token_accuracy": 0.19721650928258896,
|
|
"num_tokens": 9037628.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 6.083189868927002,
|
|
"epoch": 4.592739930382894,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003446031026637277,
|
|
"loss": 5.6061,
|
|
"mean_token_accuracy": 0.17651937007904053,
|
|
"num_tokens": 9047939.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 6.007743692398071,
|
|
"epoch": 4.597712580805569,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034423162432119995,
|
|
"loss": 5.5747,
|
|
"mean_token_accuracy": 0.18463008254766464,
|
|
"num_tokens": 9056583.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.931058692932129,
|
|
"epoch": 4.602685231228245,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003438599374109934,
|
|
"loss": 5.5363,
|
|
"mean_token_accuracy": 0.18236446380615234,
|
|
"num_tokens": 9067610.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 6.018082094192505,
|
|
"epoch": 4.60765788165092,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034348804305285466,
|
|
"loss": 5.5231,
|
|
"mean_token_accuracy": 0.17871071100234986,
|
|
"num_tokens": 9077958.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 6.015735006332397,
|
|
"epoch": 4.612630532073595,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003431159423671552,
|
|
"loss": 5.6292,
|
|
"mean_token_accuracy": 0.17197308987379073,
|
|
"num_tokens": 9088821.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.927973318099975,
|
|
"epoch": 4.61760318249627,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034274363647488826,
|
|
"loss": 5.4711,
|
|
"mean_token_accuracy": 0.1874272570014,
|
|
"num_tokens": 9098495.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.968257188796997,
|
|
"epoch": 4.622575832918946,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003423711264976652,
|
|
"loss": 5.5217,
|
|
"mean_token_accuracy": 0.18582940846681595,
|
|
"num_tokens": 9107279.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 6.015957164764404,
|
|
"epoch": 4.627548483341621,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034199841355771226,
|
|
"loss": 5.5801,
|
|
"mean_token_accuracy": 0.17783716470003127,
|
|
"num_tokens": 9118182.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.96862645149231,
|
|
"epoch": 4.632521133764296,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034162549877786685,
|
|
"loss": 5.5219,
|
|
"mean_token_accuracy": 0.18413736224174498,
|
|
"num_tokens": 9128299.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 6.052267456054688,
|
|
"epoch": 4.6374937841869714,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003412523832815748,
|
|
"loss": 5.6608,
|
|
"mean_token_accuracy": 0.16801231801509858,
|
|
"num_tokens": 9138097.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.895151090621948,
|
|
"epoch": 4.642466434609647,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003408790681928865,
|
|
"loss": 5.4133,
|
|
"mean_token_accuracy": 0.19226014316082002,
|
|
"num_tokens": 9148951.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 6.037610816955566,
|
|
"epoch": 4.647439085032322,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003405055546364535,
|
|
"loss": 5.526,
|
|
"mean_token_accuracy": 0.1839582085609436,
|
|
"num_tokens": 9158372.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.892183256149292,
|
|
"epoch": 4.652411735454997,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00034013184373752557,
|
|
"loss": 5.4001,
|
|
"mean_token_accuracy": 0.1952127069234848,
|
|
"num_tokens": 9168308.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.866609668731689,
|
|
"epoch": 4.6573843858776725,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003397579366219466,
|
|
"loss": 5.4396,
|
|
"mean_token_accuracy": 0.20242045372724532,
|
|
"num_tokens": 9178200.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.969513416290283,
|
|
"epoch": 4.662357036300348,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003393838344161519,
|
|
"loss": 5.5852,
|
|
"mean_token_accuracy": 0.1795943096280098,
|
|
"num_tokens": 9187759.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.76086745262146,
|
|
"epoch": 4.667329686723024,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033900953824716443,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.20793586522340773,
|
|
"num_tokens": 9198530.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.9109787940979,
|
|
"epoch": 4.672302337145698,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003386350492425913,
|
|
"loss": 5.5606,
|
|
"mean_token_accuracy": 0.18011290729045867,
|
|
"num_tokens": 9207668.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 6.040143251419067,
|
|
"epoch": 4.677274987568374,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00033826036853062095,
|
|
"loss": 5.6595,
|
|
"mean_token_accuracy": 0.17957742363214493,
|
|
"num_tokens": 9216856.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 6.006403970718384,
|
|
"epoch": 4.68224763799105,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033788549724001897,
|
|
"loss": 5.565,
|
|
"mean_token_accuracy": 0.17919410914182662,
|
|
"num_tokens": 9226307.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.868775177001953,
|
|
"epoch": 4.687220288413725,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033751043650012534,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.18312183171510696,
|
|
"num_tokens": 9235919.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.822739458084106,
|
|
"epoch": 4.6921929388364,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003371351874408506,
|
|
"loss": 5.2778,
|
|
"mean_token_accuracy": 0.20165799260139466,
|
|
"num_tokens": 9245542.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 6.0591672420501705,
|
|
"epoch": 4.697165589259075,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003367597511926728,
|
|
"loss": 5.6351,
|
|
"mean_token_accuracy": 0.1766924649477005,
|
|
"num_tokens": 9256146.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 6.019809722900391,
|
|
"epoch": 4.702138239681751,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033638412888663374,
|
|
"loss": 5.6793,
|
|
"mean_token_accuracy": 0.17340929210186004,
|
|
"num_tokens": 9266694.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.976730108261108,
|
|
"epoch": 4.707110890104426,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003360083216543358,
|
|
"loss": 5.5857,
|
|
"mean_token_accuracy": 0.17791337817907332,
|
|
"num_tokens": 9276155.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 6.021005630493164,
|
|
"epoch": 4.712083540527101,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003356323306279387,
|
|
"loss": 5.5631,
|
|
"mean_token_accuracy": 0.17881931513547897,
|
|
"num_tokens": 9286048.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 6.071309804916382,
|
|
"epoch": 4.717056190949776,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00033525615694015545,
|
|
"loss": 5.6649,
|
|
"mean_token_accuracy": 0.1722365766763687,
|
|
"num_tokens": 9296515.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.836947393417359,
|
|
"epoch": 4.722028841372452,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033487980172424955,
|
|
"loss": 5.4456,
|
|
"mean_token_accuracy": 0.20090975612401962,
|
|
"num_tokens": 9306737.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.989570617675781,
|
|
"epoch": 4.727001491795127,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033450326611403136,
|
|
"loss": 5.6192,
|
|
"mean_token_accuracy": 0.17521234303712846,
|
|
"num_tokens": 9316453.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.98852915763855,
|
|
"epoch": 4.731974142217802,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033412655124385494,
|
|
"loss": 5.5513,
|
|
"mean_token_accuracy": 0.18732855916023256,
|
|
"num_tokens": 9327422.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.99167275428772,
|
|
"epoch": 4.736946792640477,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003337496582486139,
|
|
"loss": 5.5232,
|
|
"mean_token_accuracy": 0.18199610561132432,
|
|
"num_tokens": 9336890.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.998954582214355,
|
|
"epoch": 4.741919443063153,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033337258826373896,
|
|
"loss": 5.5781,
|
|
"mean_token_accuracy": 0.181556136906147,
|
|
"num_tokens": 9346807.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.994000005722046,
|
|
"epoch": 4.746892093485828,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033299534242519355,
|
|
"loss": 5.5474,
|
|
"mean_token_accuracy": 0.1805335819721222,
|
|
"num_tokens": 9356343.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.99674243927002,
|
|
"epoch": 4.751864743908503,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033261792186947133,
|
|
"loss": 5.5634,
|
|
"mean_token_accuracy": 0.1777986243367195,
|
|
"num_tokens": 9366936.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 6.065751457214356,
|
|
"epoch": 4.756837394331178,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033224032773359203,
|
|
"loss": 5.5414,
|
|
"mean_token_accuracy": 0.18463751524686814,
|
|
"num_tokens": 9377520.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 6.031024217605591,
|
|
"epoch": 4.761810044753854,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033186256115509845,
|
|
"loss": 5.6547,
|
|
"mean_token_accuracy": 0.1756109908223152,
|
|
"num_tokens": 9386498.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.950701570510864,
|
|
"epoch": 4.766782695176529,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003314846232720528,
|
|
"loss": 5.5799,
|
|
"mean_token_accuracy": 0.17346807569265366,
|
|
"num_tokens": 9396153.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 6.031246852874756,
|
|
"epoch": 4.771755345599204,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033110651522303346,
|
|
"loss": 5.5896,
|
|
"mean_token_accuracy": 0.18153804391622544,
|
|
"num_tokens": 9406019.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.9756756782531735,
|
|
"epoch": 4.776727996021879,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003307282381471314,
|
|
"loss": 5.5736,
|
|
"mean_token_accuracy": 0.17805342525243759,
|
|
"num_tokens": 9416019.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.952999305725098,
|
|
"epoch": 4.781700646444555,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033034979318394676,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.18514304906129836,
|
|
"num_tokens": 9425874.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.99598445892334,
|
|
"epoch": 4.78667329686723,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00032997118147358554,
|
|
"loss": 5.5963,
|
|
"mean_token_accuracy": 0.18010542541742325,
|
|
"num_tokens": 9435252.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 6.044913578033447,
|
|
"epoch": 4.791645947289906,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000329592404156656,
|
|
"loss": 5.5482,
|
|
"mean_token_accuracy": 0.17904707044363022,
|
|
"num_tokens": 9444696.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.976610851287842,
|
|
"epoch": 4.79661859771258,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032921346237426536,
|
|
"loss": 5.57,
|
|
"mean_token_accuracy": 0.1828293174505234,
|
|
"num_tokens": 9454163.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.868893146514893,
|
|
"epoch": 4.8015912481352565,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032883435726801635,
|
|
"loss": 5.3656,
|
|
"mean_token_accuracy": 0.19810355305671692,
|
|
"num_tokens": 9463058.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.920153379440308,
|
|
"epoch": 4.806563898557932,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003284550899800036,
|
|
"loss": 5.5146,
|
|
"mean_token_accuracy": 0.1862369418144226,
|
|
"num_tokens": 9472302.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 6.074582052230835,
|
|
"epoch": 4.811536548980607,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00032807566165281055,
|
|
"loss": 5.5702,
|
|
"mean_token_accuracy": 0.17932066619396209,
|
|
"num_tokens": 9482607.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.999658584594727,
|
|
"epoch": 4.816509199403282,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003276960734295054,
|
|
"loss": 5.5576,
|
|
"mean_token_accuracy": 0.18433681577444078,
|
|
"num_tokens": 9492451.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.96454701423645,
|
|
"epoch": 4.8214818498259575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032731632645363856,
|
|
"loss": 5.603,
|
|
"mean_token_accuracy": 0.17390450537204744,
|
|
"num_tokens": 9502633.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.930258512496948,
|
|
"epoch": 4.826454500248633,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003269364218692384,
|
|
"loss": 5.4764,
|
|
"mean_token_accuracy": 0.18624778538942338,
|
|
"num_tokens": 9512887.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.928395366668701,
|
|
"epoch": 4.831427150671308,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003265563608208081,
|
|
"loss": 5.504,
|
|
"mean_token_accuracy": 0.1890804260969162,
|
|
"num_tokens": 9522210.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.8489524841308596,
|
|
"epoch": 4.836399801093983,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003261761444533221,
|
|
"loss": 5.3954,
|
|
"mean_token_accuracy": 0.1937517285346985,
|
|
"num_tokens": 9532790.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.9793782234191895,
|
|
"epoch": 4.8413724515166585,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003257957739122231,
|
|
"loss": 5.5362,
|
|
"mean_token_accuracy": 0.18291847854852678,
|
|
"num_tokens": 9543429.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 6.0334878921508786,
|
|
"epoch": 4.846345101939334,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032541525034341805,
|
|
"loss": 5.6255,
|
|
"mean_token_accuracy": 0.1756064236164093,
|
|
"num_tokens": 9552722.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.953211784362793,
|
|
"epoch": 4.851317752362009,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003250345748932749,
|
|
"loss": 5.5136,
|
|
"mean_token_accuracy": 0.18549903184175492,
|
|
"num_tokens": 9560880.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 6.022203016281128,
|
|
"epoch": 4.856290402784684,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032465374870861913,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.1760057106614113,
|
|
"num_tokens": 9570553.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.9256126403808596,
|
|
"epoch": 4.8612630532073595,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032427277293673053,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.1743432253599167,
|
|
"num_tokens": 9581313.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.998179912567139,
|
|
"epoch": 4.866235703630035,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003238916487253393,
|
|
"loss": 5.55,
|
|
"mean_token_accuracy": 0.18060620427131652,
|
|
"num_tokens": 9589545.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 6.0333093166351315,
|
|
"epoch": 4.87120835405271,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003235103772226229,
|
|
"loss": 5.5984,
|
|
"mean_token_accuracy": 0.17442011535167695,
|
|
"num_tokens": 9599489.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.950501203536987,
|
|
"epoch": 4.876181004475385,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003231289595772026,
|
|
"loss": 5.6632,
|
|
"mean_token_accuracy": 0.17879505157470704,
|
|
"num_tokens": 9609346.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.773447465896607,
|
|
"epoch": 4.8811536548980605,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003227473969381398,
|
|
"loss": 5.3,
|
|
"mean_token_accuracy": 0.20836496502161025,
|
|
"num_tokens": 9618808.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 6.006798124313354,
|
|
"epoch": 4.886126305320736,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032236569045493283,
|
|
"loss": 5.4978,
|
|
"mean_token_accuracy": 0.18228981643915176,
|
|
"num_tokens": 9629312.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 6.035969400405884,
|
|
"epoch": 4.891098955743411,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032198384127751344,
|
|
"loss": 5.6989,
|
|
"mean_token_accuracy": 0.17527570575475693,
|
|
"num_tokens": 9639439.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 6.026311588287354,
|
|
"epoch": 4.896071606166086,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000321601850556243,
|
|
"loss": 5.5193,
|
|
"mean_token_accuracy": 0.18186003118753433,
|
|
"num_tokens": 9647824.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.993879508972168,
|
|
"epoch": 4.9010442565887615,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003212197194419095,
|
|
"loss": 5.5428,
|
|
"mean_token_accuracy": 0.18316387236118317,
|
|
"num_tokens": 9656962.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.944897079467774,
|
|
"epoch": 4.906016907011437,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003208374490857238,
|
|
"loss": 5.6347,
|
|
"mean_token_accuracy": 0.17825938612222672,
|
|
"num_tokens": 9666360.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 6.002035760879517,
|
|
"epoch": 4.910989557434112,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003204550406393164,
|
|
"loss": 5.6163,
|
|
"mean_token_accuracy": 0.17632779777050017,
|
|
"num_tokens": 9676355.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 6.052821683883667,
|
|
"epoch": 4.915962207856787,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003200724952547335,
|
|
"loss": 5.6058,
|
|
"mean_token_accuracy": 0.17976820468902588,
|
|
"num_tokens": 9686361.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.963199758529663,
|
|
"epoch": 4.9209348582794625,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003196898140844341,
|
|
"loss": 5.5105,
|
|
"mean_token_accuracy": 0.1862466812133789,
|
|
"num_tokens": 9696571.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 6.079383373260498,
|
|
"epoch": 4.925907508702139,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031930699828128614,
|
|
"loss": 5.7183,
|
|
"mean_token_accuracy": 0.17418165877461433,
|
|
"num_tokens": 9707595.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.992888355255127,
|
|
"epoch": 4.930880159124813,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003189240489985633,
|
|
"loss": 5.5969,
|
|
"mean_token_accuracy": 0.1831539034843445,
|
|
"num_tokens": 9717094.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.946429920196533,
|
|
"epoch": 4.935852809547489,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003185409673899412,
|
|
"loss": 5.538,
|
|
"mean_token_accuracy": 0.18024692684412003,
|
|
"num_tokens": 9727329.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.903452777862549,
|
|
"epoch": 4.940825459970164,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00031815775460949426,
|
|
"loss": 5.4217,
|
|
"mean_token_accuracy": 0.1873863399028778,
|
|
"num_tokens": 9736189.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.99541745185852,
|
|
"epoch": 4.94579811039284,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003177744118116919,
|
|
"loss": 5.5142,
|
|
"mean_token_accuracy": 0.18631737381219865,
|
|
"num_tokens": 9745760.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 6.111676454544067,
|
|
"epoch": 4.950770760815515,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003173909401513954,
|
|
"loss": 5.6526,
|
|
"mean_token_accuracy": 0.16547196507453918,
|
|
"num_tokens": 9755941.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.970871067047119,
|
|
"epoch": 4.95574341123819,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00031700734078385416,
|
|
"loss": 5.5167,
|
|
"mean_token_accuracy": 0.18113433569669724,
|
|
"num_tokens": 9764546.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 6.041245222091675,
|
|
"epoch": 4.960716061660865,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031662361486470243,
|
|
"loss": 5.5747,
|
|
"mean_token_accuracy": 0.17844264656305314,
|
|
"num_tokens": 9773728.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.9237682819366455,
|
|
"epoch": 4.965688712083541,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003162397635499555,
|
|
"loss": 5.5444,
|
|
"mean_token_accuracy": 0.17766229957342147,
|
|
"num_tokens": 9783878.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.967093133926392,
|
|
"epoch": 4.970661362506216,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00031585578799600653,
|
|
"loss": 5.573,
|
|
"mean_token_accuracy": 0.17680271118879318,
|
|
"num_tokens": 9793926.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.970661362506216,
|
|
"eval_entropy": 5.914577451169765,
|
|
"eval_loss": 6.530643463134766,
|
|
"eval_mean_token_accuracy": 0.14372347328760382,
|
|
"eval_num_tokens": 9793926.0,
|
|
"eval_runtime": 10.4851,
|
|
"eval_samples_per_second": 330.469,
|
|
"eval_steps_per_second": 41.392,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.973652458190918,
|
|
"epoch": 4.975634012928891,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031547168935962304,
|
|
"loss": 5.4964,
|
|
"mean_token_accuracy": 0.19288804978132248,
|
|
"num_tokens": 9804213.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.957664632797242,
|
|
"epoch": 4.980606663351566,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003150874687979433,
|
|
"loss": 5.5451,
|
|
"mean_token_accuracy": 0.18217722922563553,
|
|
"num_tokens": 9814151.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.972400665283203,
|
|
"epoch": 4.985579313774242,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031470312746847284,
|
|
"loss": 5.5437,
|
|
"mean_token_accuracy": 0.1827799528837204,
|
|
"num_tokens": 9822646.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.849236917495728,
|
|
"epoch": 4.990551964196917,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031431866652908107,
|
|
"loss": 5.3186,
|
|
"mean_token_accuracy": 0.20924449861049652,
|
|
"num_tokens": 9833204.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.853102445602417,
|
|
"epoch": 4.995524614619592,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003139340871379977,
|
|
"loss": 5.4397,
|
|
"mean_token_accuracy": 0.20606417804956437,
|
|
"num_tokens": 9843150.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.972547001308865,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 0.0003135493904538093,
|
|
"loss": 5.5062,
|
|
"mean_token_accuracy": 0.18399477501710257,
|
|
"num_tokens": 9850520.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.982021999359131,
|
|
"epoch": 5.004972650422675,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003131645776354559,
|
|
"loss": 5.2615,
|
|
"mean_token_accuracy": 0.19307033121585845,
|
|
"num_tokens": 9860876.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.878837585449219,
|
|
"epoch": 5.0099453008453505,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003127796498422272,
|
|
"loss": 5.1569,
|
|
"mean_token_accuracy": 0.21490370631217956,
|
|
"num_tokens": 9870944.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.920594978332519,
|
|
"epoch": 5.014917951268026,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031239460823375933,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.2005557596683502,
|
|
"num_tokens": 9880615.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.766988706588745,
|
|
"epoch": 5.019890601690701,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031200945397003157,
|
|
"loss": 5.0421,
|
|
"mean_token_accuracy": 0.22147153317928314,
|
|
"num_tokens": 9889933.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 6.005528354644776,
|
|
"epoch": 5.024863252113376,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003116241882113623,
|
|
"loss": 5.32,
|
|
"mean_token_accuracy": 0.1874865099787712,
|
|
"num_tokens": 9900166.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.86612229347229,
|
|
"epoch": 5.0298359025360515,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00031123881211840565,
|
|
"loss": 5.2211,
|
|
"mean_token_accuracy": 0.2021344542503357,
|
|
"num_tokens": 9909747.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.8963525772094725,
|
|
"epoch": 5.034808552958727,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003108533268521485,
|
|
"loss": 5.2625,
|
|
"mean_token_accuracy": 0.2018459305167198,
|
|
"num_tokens": 9919039.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.967410278320313,
|
|
"epoch": 5.039781203381402,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003104677335739065,
|
|
"loss": 5.2812,
|
|
"mean_token_accuracy": 0.2000605508685112,
|
|
"num_tokens": 9927933.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.961862516403198,
|
|
"epoch": 5.044753853804077,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031008203344532067,
|
|
"loss": 5.258,
|
|
"mean_token_accuracy": 0.19686445742845535,
|
|
"num_tokens": 9939264.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.933650350570678,
|
|
"epoch": 5.0497265042267525,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000309696227628354,
|
|
"loss": 5.294,
|
|
"mean_token_accuracy": 0.19528610706329347,
|
|
"num_tokens": 9948375.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.919182157516479,
|
|
"epoch": 5.054699154649428,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00030931031728528754,
|
|
"loss": 5.1788,
|
|
"mean_token_accuracy": 0.20404241979122162,
|
|
"num_tokens": 9957792.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.919578742980957,
|
|
"epoch": 5.059671805072103,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003089243035787178,
|
|
"loss": 5.2478,
|
|
"mean_token_accuracy": 0.20509056448936464,
|
|
"num_tokens": 9967213.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.907741928100586,
|
|
"epoch": 5.064644455494779,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030853818767155225,
|
|
"loss": 5.2974,
|
|
"mean_token_accuracy": 0.1938672348856926,
|
|
"num_tokens": 9976886.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.882480096817017,
|
|
"epoch": 5.069617105917454,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003081519707270065,
|
|
"loss": 5.214,
|
|
"mean_token_accuracy": 0.20390648692846297,
|
|
"num_tokens": 9986702.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.948557949066162,
|
|
"epoch": 5.07458975634013,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030776565390860017,
|
|
"loss": 5.1957,
|
|
"mean_token_accuracy": 0.2075501039624214,
|
|
"num_tokens": 9996597.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.895875930786133,
|
|
"epoch": 5.079562406762805,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003073792383801543,
|
|
"loss": 5.278,
|
|
"mean_token_accuracy": 0.19709229320287705,
|
|
"num_tokens": 10006440.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.923911905288696,
|
|
"epoch": 5.08453505718548,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00030699272530578697,
|
|
"loss": 5.2834,
|
|
"mean_token_accuracy": 0.19854919910430907,
|
|
"num_tokens": 10016198.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.980761909484864,
|
|
"epoch": 5.089507707608155,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030660611584991013,
|
|
"loss": 5.3033,
|
|
"mean_token_accuracy": 0.19637626856565477,
|
|
"num_tokens": 10025795.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.889021110534668,
|
|
"epoch": 5.094480358030831,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00030621941117722615,
|
|
"loss": 5.1852,
|
|
"mean_token_accuracy": 0.2104475811123848,
|
|
"num_tokens": 10035448.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.849649953842163,
|
|
"epoch": 5.099453008453506,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003058326124527243,
|
|
"loss": 5.2738,
|
|
"mean_token_accuracy": 0.20311955809593202,
|
|
"num_tokens": 10046335.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.993789768218994,
|
|
"epoch": 5.104425658876181,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003054457208416771,
|
|
"loss": 5.329,
|
|
"mean_token_accuracy": 0.19182431548833848,
|
|
"num_tokens": 10056346.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.788894367218018,
|
|
"epoch": 5.109398309298856,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030505873750963705,
|
|
"loss": 5.1225,
|
|
"mean_token_accuracy": 0.21736940294504165,
|
|
"num_tokens": 10065948.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.842388725280761,
|
|
"epoch": 5.114370959721532,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0003046716636224327,
|
|
"loss": 5.2043,
|
|
"mean_token_accuracy": 0.20182041972875595,
|
|
"num_tokens": 10076680.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.9237611293792725,
|
|
"epoch": 5.119343610144207,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003042845003461657,
|
|
"loss": 5.2693,
|
|
"mean_token_accuracy": 0.1926799401640892,
|
|
"num_tokens": 10086373.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.9351307392120365,
|
|
"epoch": 5.124316260566882,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030389724884720664,
|
|
"loss": 5.2475,
|
|
"mean_token_accuracy": 0.1956130623817444,
|
|
"num_tokens": 10095415.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.736910009384156,
|
|
"epoch": 5.129288910989557,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030350991029219236,
|
|
"loss": 5.1135,
|
|
"mean_token_accuracy": 0.22079250663518907,
|
|
"num_tokens": 10104709.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.9395928382873535,
|
|
"epoch": 5.134261561412233,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00030312248584802166,
|
|
"loss": 5.2658,
|
|
"mean_token_accuracy": 0.197672276198864,
|
|
"num_tokens": 10113539.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.783051538467407,
|
|
"epoch": 5.139234211834908,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003027349766818521,
|
|
"loss": 5.1466,
|
|
"mean_token_accuracy": 0.21628278195858003,
|
|
"num_tokens": 10123914.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.964668941497803,
|
|
"epoch": 5.144206862257583,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003023473839610965,
|
|
"loss": 5.3236,
|
|
"mean_token_accuracy": 0.19525672346353531,
|
|
"num_tokens": 10133572.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.969754648208618,
|
|
"epoch": 5.149179512680258,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003019597088534195,
|
|
"loss": 5.3703,
|
|
"mean_token_accuracy": 0.19450667649507522,
|
|
"num_tokens": 10142973.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.959574222564697,
|
|
"epoch": 5.154152163102934,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003015719525267338,
|
|
"loss": 5.354,
|
|
"mean_token_accuracy": 0.1932512789964676,
|
|
"num_tokens": 10153322.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.893978452682495,
|
|
"epoch": 5.159124813525609,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003011841161491969,
|
|
"loss": 5.2431,
|
|
"mean_token_accuracy": 0.20005542784929276,
|
|
"num_tokens": 10163176.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.9512755393981935,
|
|
"epoch": 5.164097463948284,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030079620088920727,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.1974889039993286,
|
|
"num_tokens": 10172616.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.9298789501190186,
|
|
"epoch": 5.169070114370959,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003004082079154013,
|
|
"loss": 5.3038,
|
|
"mean_token_accuracy": 0.19613510221242905,
|
|
"num_tokens": 10182243.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.9262457370758055,
|
|
"epoch": 5.174042764793635,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030002013839664925,
|
|
"loss": 5.2634,
|
|
"mean_token_accuracy": 0.19784385412931443,
|
|
"num_tokens": 10191394.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.704701042175293,
|
|
"epoch": 5.17901541521631,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002996319935020522,
|
|
"loss": 5.116,
|
|
"mean_token_accuracy": 0.22243674993515014,
|
|
"num_tokens": 10202167.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.878773927688599,
|
|
"epoch": 5.183988065638985,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002992437744009381,
|
|
"loss": 5.1714,
|
|
"mean_token_accuracy": 0.20399222224950792,
|
|
"num_tokens": 10211250.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.96918888092041,
|
|
"epoch": 5.188960716061661,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029885548226285855,
|
|
"loss": 5.3441,
|
|
"mean_token_accuracy": 0.1949370965361595,
|
|
"num_tokens": 10220722.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.959464740753174,
|
|
"epoch": 5.193933366484337,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002984671182575851,
|
|
"loss": 5.3375,
|
|
"mean_token_accuracy": 0.19210466295480727,
|
|
"num_tokens": 10229880.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.911393642425537,
|
|
"epoch": 5.198906016907012,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002980786835551061,
|
|
"loss": 5.328,
|
|
"mean_token_accuracy": 0.19600168913602828,
|
|
"num_tokens": 10240939.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.960146856307984,
|
|
"epoch": 5.203878667329687,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002976901793256225,
|
|
"loss": 5.285,
|
|
"mean_token_accuracy": 0.19698677211999893,
|
|
"num_tokens": 10249383.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.9890944480896,
|
|
"epoch": 5.208851317752362,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000297301606739545,
|
|
"loss": 5.3256,
|
|
"mean_token_accuracy": 0.19104150384664537,
|
|
"num_tokens": 10259166.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.970776891708374,
|
|
"epoch": 5.213823968175038,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00029691296696748994,
|
|
"loss": 5.4001,
|
|
"mean_token_accuracy": 0.1885468289256096,
|
|
"num_tokens": 10269053.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.92987470626831,
|
|
"epoch": 5.218796618597713,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029652426118027635,
|
|
"loss": 5.2814,
|
|
"mean_token_accuracy": 0.20062332153320311,
|
|
"num_tokens": 10279148.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.861099529266357,
|
|
"epoch": 5.223769269020388,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00029613549054892206,
|
|
"loss": 5.2334,
|
|
"mean_token_accuracy": 0.19522044509649278,
|
|
"num_tokens": 10288719.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.87611756324768,
|
|
"epoch": 5.228741919443063,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002957466562446401,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.1991399809718132,
|
|
"num_tokens": 10300089.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.937896680831909,
|
|
"epoch": 5.233714569865739,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002953577594388355,
|
|
"loss": 5.262,
|
|
"mean_token_accuracy": 0.1958780974149704,
|
|
"num_tokens": 10309400.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.940013837814331,
|
|
"epoch": 5.238687220288414,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00029496880130310145,
|
|
"loss": 5.3199,
|
|
"mean_token_accuracy": 0.20040978193283082,
|
|
"num_tokens": 10319115.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.908069896697998,
|
|
"epoch": 5.243659870711089,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002945797830092161,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.2027330994606018,
|
|
"num_tokens": 10327174.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.939208745956421,
|
|
"epoch": 5.248632521133764,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002941907057291386,
|
|
"loss": 5.2946,
|
|
"mean_token_accuracy": 0.20115703344345093,
|
|
"num_tokens": 10335608.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.8390120506286625,
|
|
"epoch": 5.25360517155644,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029380157063500594,
|
|
"loss": 5.2109,
|
|
"mean_token_accuracy": 0.20305149108171464,
|
|
"num_tokens": 10345592.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.815340948104859,
|
|
"epoch": 5.258577821979115,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002934123788991291,
|
|
"loss": 5.2201,
|
|
"mean_token_accuracy": 0.2067882314324379,
|
|
"num_tokens": 10355913.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.913070964813232,
|
|
"epoch": 5.26355047240179,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00029302313169399,
|
|
"loss": 5.3079,
|
|
"mean_token_accuracy": 0.1991267278790474,
|
|
"num_tokens": 10365419.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.948590993881226,
|
|
"epoch": 5.268523122824465,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002926338301922376,
|
|
"loss": 5.3033,
|
|
"mean_token_accuracy": 0.1916865512728691,
|
|
"num_tokens": 10373939.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.949945259094238,
|
|
"epoch": 5.273495773247141,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00029224447556668417,
|
|
"loss": 5.341,
|
|
"mean_token_accuracy": 0.1931505635380745,
|
|
"num_tokens": 10383932.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.954394578933716,
|
|
"epoch": 5.278468423669816,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002918550689903022,
|
|
"loss": 5.2683,
|
|
"mean_token_accuracy": 0.19998514354228974,
|
|
"num_tokens": 10393319.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.897801733016967,
|
|
"epoch": 5.283441074092491,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029146561163622076,
|
|
"loss": 5.4138,
|
|
"mean_token_accuracy": 0.18264214992523192,
|
|
"num_tokens": 10404102.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.855690956115723,
|
|
"epoch": 5.288413724515166,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002910761046777218,
|
|
"loss": 5.2748,
|
|
"mean_token_accuracy": 0.19668395072221756,
|
|
"num_tokens": 10412718.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.972980785369873,
|
|
"epoch": 5.293386374937842,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00029068654928823666,
|
|
"loss": 5.3548,
|
|
"mean_token_accuracy": 0.19323400110006334,
|
|
"num_tokens": 10422546.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.953373003005981,
|
|
"epoch": 5.298359025360517,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029029694664134284,
|
|
"loss": 5.3288,
|
|
"mean_token_accuracy": 0.19082262814044954,
|
|
"num_tokens": 10431498.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.956353569030762,
|
|
"epoch": 5.303331675783192,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002899072979107597,
|
|
"loss": 5.3336,
|
|
"mean_token_accuracy": 0.19732917994260787,
|
|
"num_tokens": 10441713.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.99358606338501,
|
|
"epoch": 5.308304326205867,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002895176042703461,
|
|
"loss": 5.3502,
|
|
"mean_token_accuracy": 0.18930336833000183,
|
|
"num_tokens": 10451570.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.9613348007202145,
|
|
"epoch": 5.3132769766285435,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028912786689409554,
|
|
"loss": 5.3162,
|
|
"mean_token_accuracy": 0.19836218804121017,
|
|
"num_tokens": 10461442.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.997992944717407,
|
|
"epoch": 5.318249627051218,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002887380869561338,
|
|
"loss": 5.3896,
|
|
"mean_token_accuracy": 0.18766831159591674,
|
|
"num_tokens": 10471223.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.884400415420532,
|
|
"epoch": 5.323222277473894,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002883482656307145,
|
|
"loss": 5.25,
|
|
"mean_token_accuracy": 0.194474658370018,
|
|
"num_tokens": 10480469.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.859659862518311,
|
|
"epoch": 5.328194927896569,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002879584040922162,
|
|
"loss": 5.3043,
|
|
"mean_token_accuracy": 0.19395188242197037,
|
|
"num_tokens": 10490320.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.847450399398804,
|
|
"epoch": 5.3331675783192445,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002875685035151386,
|
|
"loss": 5.3437,
|
|
"mean_token_accuracy": 0.18698538541793824,
|
|
"num_tokens": 10500471.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 6.015448999404907,
|
|
"epoch": 5.33814022874192,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00028717856507409886,
|
|
"loss": 5.4098,
|
|
"mean_token_accuracy": 0.18574409186840057,
|
|
"num_tokens": 10510956.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 6.021253776550293,
|
|
"epoch": 5.343112879164595,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002867885899438281,
|
|
"loss": 5.3899,
|
|
"mean_token_accuracy": 0.19838718175888062,
|
|
"num_tokens": 10520838.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.9923032283782955,
|
|
"epoch": 5.34808552958727,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002863985792991683,
|
|
"loss": 5.3267,
|
|
"mean_token_accuracy": 0.19674330949783325,
|
|
"num_tokens": 10530879.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.77642183303833,
|
|
"epoch": 5.3530581800099455,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002860085343150683,
|
|
"loss": 5.1298,
|
|
"mean_token_accuracy": 0.22051493227481841,
|
|
"num_tokens": 10540161.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.965254497528076,
|
|
"epoch": 5.358030830432621,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00028561845616658033,
|
|
"loss": 5.3421,
|
|
"mean_token_accuracy": 0.1922277733683586,
|
|
"num_tokens": 10551173.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.929450559616089,
|
|
"epoch": 5.363003480855296,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028522834602885656,
|
|
"loss": 5.2867,
|
|
"mean_token_accuracy": 0.19767429828643798,
|
|
"num_tokens": 10560882.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.988594198226929,
|
|
"epoch": 5.367976131277971,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002848382050771455,
|
|
"loss": 5.4239,
|
|
"mean_token_accuracy": 0.1898815244436264,
|
|
"num_tokens": 10570837.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.947837018966675,
|
|
"epoch": 5.3729487817006465,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002844480344867886,
|
|
"loss": 5.3117,
|
|
"mean_token_accuracy": 0.20116714388132095,
|
|
"num_tokens": 10580335.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.913256978988647,
|
|
"epoch": 5.377921432123322,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002840578354332165,
|
|
"loss": 5.3142,
|
|
"mean_token_accuracy": 0.19572788327932358,
|
|
"num_tokens": 10589030.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.910342121124268,
|
|
"epoch": 5.382894082545997,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002836676090919456,
|
|
"loss": 5.3011,
|
|
"mean_token_accuracy": 0.19332673102617265,
|
|
"num_tokens": 10599309.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.927676868438721,
|
|
"epoch": 5.387866732968672,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002832773566385746,
|
|
"loss": 5.2731,
|
|
"mean_token_accuracy": 0.20543116629123687,
|
|
"num_tokens": 10608781.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.862621879577636,
|
|
"epoch": 5.3928393833913475,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00028288707924878056,
|
|
"loss": 5.2046,
|
|
"mean_token_accuracy": 0.20060649365186692,
|
|
"num_tokens": 10618512.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.946128177642822,
|
|
"epoch": 5.397812033814023,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002824967780983162,
|
|
"loss": 5.3371,
|
|
"mean_token_accuracy": 0.19224245101213455,
|
|
"num_tokens": 10629420.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.952072381973267,
|
|
"epoch": 5.402784684236698,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028210645436300524,
|
|
"loss": 5.321,
|
|
"mean_token_accuracy": 0.19605873078107833,
|
|
"num_tokens": 10639625.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.844250345230103,
|
|
"epoch": 5.407757334659373,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00028171610921873995,
|
|
"loss": 5.225,
|
|
"mean_token_accuracy": 0.1961871787905693,
|
|
"num_tokens": 10648633.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.7801251888275145,
|
|
"epoch": 5.4127299850820485,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002813257438414766,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.20574670732021333,
|
|
"num_tokens": 10657981.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.83755259513855,
|
|
"epoch": 5.417702635504724,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000280935359407233,
|
|
"loss": 5.1673,
|
|
"mean_token_accuracy": 0.20683670043945312,
|
|
"num_tokens": 10669734.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.924308347702026,
|
|
"epoch": 5.422675285927399,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002805449570920838,
|
|
"loss": 5.283,
|
|
"mean_token_accuracy": 0.19454064220190048,
|
|
"num_tokens": 10679486.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.896684551239014,
|
|
"epoch": 5.427647936350074,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00028015453807215786,
|
|
"loss": 5.3232,
|
|
"mean_token_accuracy": 0.19298199862241744,
|
|
"num_tokens": 10689551.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.889700269699096,
|
|
"epoch": 5.4326205867727495,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027976410352363416,
|
|
"loss": 5.2871,
|
|
"mean_token_accuracy": 0.1911845698952675,
|
|
"num_tokens": 10699173.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.855425691604614,
|
|
"epoch": 5.437593237195425,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027937365462273864,
|
|
"loss": 5.2485,
|
|
"mean_token_accuracy": 0.19973269253969192,
|
|
"num_tokens": 10709050.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.897402811050415,
|
|
"epoch": 5.4425658876181,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027898319254574043,
|
|
"loss": 5.3281,
|
|
"mean_token_accuracy": 0.19833158403635026,
|
|
"num_tokens": 10718397.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.915306949615479,
|
|
"epoch": 5.447538538040776,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002785927184689483,
|
|
"loss": 5.368,
|
|
"mean_token_accuracy": 0.19333116263151168,
|
|
"num_tokens": 10729305.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.934562253952026,
|
|
"epoch": 5.452511188463451,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002782022335687073,
|
|
"loss": 5.2834,
|
|
"mean_token_accuracy": 0.19437097907066345,
|
|
"num_tokens": 10738602.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.876776123046875,
|
|
"epoch": 5.457483838886127,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00027781173902139485,
|
|
"loss": 5.2874,
|
|
"mean_token_accuracy": 0.2095681130886078,
|
|
"num_tokens": 10748373.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.907406234741211,
|
|
"epoch": 5.462456489308802,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002774212360034176,
|
|
"loss": 5.3234,
|
|
"mean_token_accuracy": 0.19256188869476318,
|
|
"num_tokens": 10756879.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.855845594406128,
|
|
"epoch": 5.467429139731477,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002770307256912078,
|
|
"loss": 5.1777,
|
|
"mean_token_accuracy": 0.20358728468418122,
|
|
"num_tokens": 10766969.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 5.467429139731477,
|
|
"eval_entropy": 5.839827089265745,
|
|
"eval_loss": 6.543715000152588,
|
|
"eval_mean_token_accuracy": 0.1447248482553091,
|
|
"eval_num_tokens": 10766969.0,
|
|
"eval_runtime": 15.0415,
|
|
"eval_samples_per_second": 230.363,
|
|
"eval_steps_per_second": 28.854,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.939993953704834,
|
|
"epoch": 5.472401790154152,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027664020926121944,
|
|
"loss": 5.3252,
|
|
"mean_token_accuracy": 0.19736808091402053,
|
|
"num_tokens": 10776780.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.910236406326294,
|
|
"epoch": 5.477374440576828,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002762496878899251,
|
|
"loss": 5.341,
|
|
"mean_token_accuracy": 0.19337387830018998,
|
|
"num_tokens": 10786341.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.926927518844605,
|
|
"epoch": 5.482347090999503,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027585916275381226,
|
|
"loss": 5.2377,
|
|
"mean_token_accuracy": 0.2030269980430603,
|
|
"num_tokens": 10795754.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.91486964225769,
|
|
"epoch": 5.487319741422178,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002754686350293797,
|
|
"loss": 5.4089,
|
|
"mean_token_accuracy": 0.1878177762031555,
|
|
"num_tokens": 10805796.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.841636323928833,
|
|
"epoch": 5.492292391844853,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002750781058931339,
|
|
"loss": 5.146,
|
|
"mean_token_accuracy": 0.21309227198362352,
|
|
"num_tokens": 10815427.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.917548370361328,
|
|
"epoch": 5.497265042267529,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002746875765215857,
|
|
"loss": 5.2735,
|
|
"mean_token_accuracy": 0.2015067905187607,
|
|
"num_tokens": 10825501.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.974257612228394,
|
|
"epoch": 5.502237692690204,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00027429704809124665,
|
|
"loss": 5.397,
|
|
"mean_token_accuracy": 0.19115436673164368,
|
|
"num_tokens": 10834961.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.941097736358643,
|
|
"epoch": 5.507210343112879,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002739065217786255,
|
|
"loss": 5.31,
|
|
"mean_token_accuracy": 0.19461524486541748,
|
|
"num_tokens": 10845491.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.881209754943848,
|
|
"epoch": 5.512182993535554,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002735159987602246,
|
|
"loss": 5.3563,
|
|
"mean_token_accuracy": 0.19303604513406752,
|
|
"num_tokens": 10855045.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.899763679504394,
|
|
"epoch": 5.51715564395823,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002731254802125362,
|
|
"loss": 5.403,
|
|
"mean_token_accuracy": 0.18474946767091752,
|
|
"num_tokens": 10864254.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.9446234703063965,
|
|
"epoch": 5.522128294380905,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00027273496731203945,
|
|
"loss": 5.3033,
|
|
"mean_token_accuracy": 0.1989415168762207,
|
|
"num_tokens": 10874768.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.918610095977783,
|
|
"epoch": 5.52710094480358,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000272344461235196,
|
|
"loss": 5.2664,
|
|
"mean_token_accuracy": 0.20018296241760253,
|
|
"num_tokens": 10884820.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.751319885253906,
|
|
"epoch": 5.532073595226255,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00027195396315844733,
|
|
"loss": 5.1833,
|
|
"mean_token_accuracy": 0.21568529605865477,
|
|
"num_tokens": 10895393.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.900951862335205,
|
|
"epoch": 5.537046245648931,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002715634742582107,
|
|
"loss": 5.2195,
|
|
"mean_token_accuracy": 0.20461428463459014,
|
|
"num_tokens": 10903875.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.95953369140625,
|
|
"epoch": 5.542018896071606,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027117299571087576,
|
|
"loss": 5.371,
|
|
"mean_token_accuracy": 0.18854617923498154,
|
|
"num_tokens": 10913861.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.888231372833252,
|
|
"epoch": 5.546991546494281,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002707825286928008,
|
|
"loss": 5.3518,
|
|
"mean_token_accuracy": 0.18709164708852768,
|
|
"num_tokens": 10923597.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.8739758968353275,
|
|
"epoch": 5.551964196916956,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002703920743803097,
|
|
"loss": 5.3528,
|
|
"mean_token_accuracy": 0.1861886128783226,
|
|
"num_tokens": 10934094.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.998759317398071,
|
|
"epoch": 5.556936847339632,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00027000163394968763,
|
|
"loss": 5.425,
|
|
"mean_token_accuracy": 0.18230110704898833,
|
|
"num_tokens": 10944121.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.876296186447144,
|
|
"epoch": 5.561909497762308,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026961120857717843,
|
|
"loss": 5.2309,
|
|
"mean_token_accuracy": 0.21068502962589264,
|
|
"num_tokens": 10955117.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.792949676513672,
|
|
"epoch": 5.566882148184982,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026922079943898016,
|
|
"loss": 5.1963,
|
|
"mean_token_accuracy": 0.211909157037735,
|
|
"num_tokens": 10965248.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.971052265167236,
|
|
"epoch": 5.571854798607658,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002688304077112421,
|
|
"loss": 5.3954,
|
|
"mean_token_accuracy": 0.19308123290538787,
|
|
"num_tokens": 10975610.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.8863600730896,
|
|
"epoch": 5.576827449030333,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002684400345700613,
|
|
"loss": 5.266,
|
|
"mean_token_accuracy": 0.20242853015661239,
|
|
"num_tokens": 10984421.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.776471471786499,
|
|
"epoch": 5.581800099453009,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002680496811914786,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.21088954508304597,
|
|
"num_tokens": 10993740.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.856486368179321,
|
|
"epoch": 5.586772749875684,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00026765934875147513,
|
|
"loss": 5.2188,
|
|
"mean_token_accuracy": 0.20176612138748168,
|
|
"num_tokens": 11003382.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.886099863052368,
|
|
"epoch": 5.591745400298359,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002672690384259693,
|
|
"loss": 5.2413,
|
|
"mean_token_accuracy": 0.20804143697023392,
|
|
"num_tokens": 11013336.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.911739683151245,
|
|
"epoch": 5.596718050721035,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00026687875139081287,
|
|
"loss": 5.3234,
|
|
"mean_token_accuracy": 0.19484509378671647,
|
|
"num_tokens": 11023800.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 6.003859281539917,
|
|
"epoch": 5.60169070114371,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026648848882178703,
|
|
"loss": 5.501,
|
|
"mean_token_accuracy": 0.18471099883317948,
|
|
"num_tokens": 11033250.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.8714100360870365,
|
|
"epoch": 5.606663351566385,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002660982518945998,
|
|
"loss": 5.2829,
|
|
"mean_token_accuracy": 0.19726954102516175,
|
|
"num_tokens": 11042045.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.839722967147827,
|
|
"epoch": 5.61163600198906,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026570804178488146,
|
|
"loss": 5.2275,
|
|
"mean_token_accuracy": 0.20788590461015702,
|
|
"num_tokens": 11051349.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.903298377990723,
|
|
"epoch": 5.616608652411736,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002653178596681819,
|
|
"loss": 5.2785,
|
|
"mean_token_accuracy": 0.19645078778266906,
|
|
"num_tokens": 11061550.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.86050443649292,
|
|
"epoch": 5.621581302834411,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002649277067199665,
|
|
"loss": 5.2861,
|
|
"mean_token_accuracy": 0.1959319144487381,
|
|
"num_tokens": 11071887.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.810476446151734,
|
|
"epoch": 5.626553953257086,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002645375841156127,
|
|
"loss": 5.2068,
|
|
"mean_token_accuracy": 0.2034299373626709,
|
|
"num_tokens": 11081276.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.95744948387146,
|
|
"epoch": 5.631526603679761,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002641474930304067,
|
|
"loss": 5.3784,
|
|
"mean_token_accuracy": 0.18759866654872895,
|
|
"num_tokens": 11091761.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.788309335708618,
|
|
"epoch": 5.636499254102437,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002637574346395396,
|
|
"loss": 5.0972,
|
|
"mean_token_accuracy": 0.21790954768657683,
|
|
"num_tokens": 11101986.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.9237480640411375,
|
|
"epoch": 5.641471904525112,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000263367410118104,
|
|
"loss": 5.3073,
|
|
"mean_token_accuracy": 0.2000094994902611,
|
|
"num_tokens": 11111835.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.834815120697021,
|
|
"epoch": 5.646444554947787,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002629774206410906,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.2024150788784027,
|
|
"num_tokens": 11122646.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.972793626785278,
|
|
"epoch": 5.651417205370462,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002625874673833844,
|
|
"loss": 5.3164,
|
|
"mean_token_accuracy": 0.19532543420791626,
|
|
"num_tokens": 11132693.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.944497871398926,
|
|
"epoch": 5.656389855793138,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002621975515197613,
|
|
"loss": 5.3638,
|
|
"mean_token_accuracy": 0.1863997384905815,
|
|
"num_tokens": 11142285.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.876307821273803,
|
|
"epoch": 5.661362506215813,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026180767422488466,
|
|
"loss": 5.284,
|
|
"mean_token_accuracy": 0.19788937419652938,
|
|
"num_tokens": 11150657.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.875021076202392,
|
|
"epoch": 5.666335156638488,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026141783667330156,
|
|
"loss": 5.3107,
|
|
"mean_token_accuracy": 0.2010410949587822,
|
|
"num_tokens": 11160391.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.974457120895385,
|
|
"epoch": 5.671307807061163,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002610280400394393,
|
|
"loss": 5.467,
|
|
"mean_token_accuracy": 0.19171942472457887,
|
|
"num_tokens": 11169919.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.939026546478272,
|
|
"epoch": 5.676280457483839,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026063828549760195,
|
|
"loss": 5.3443,
|
|
"mean_token_accuracy": 0.19224839210510253,
|
|
"num_tokens": 11179948.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.775281810760498,
|
|
"epoch": 5.681253107906514,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002602485742219667,
|
|
"loss": 5.1055,
|
|
"mean_token_accuracy": 0.22228876799345015,
|
|
"num_tokens": 11190875.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.791461753845215,
|
|
"epoch": 5.686225758329189,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00025985890738658077,
|
|
"loss": 5.1273,
|
|
"mean_token_accuracy": 0.2178652212023735,
|
|
"num_tokens": 11201134.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.874873208999634,
|
|
"epoch": 5.691198408751864,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000259469286165357,
|
|
"loss": 5.2266,
|
|
"mean_token_accuracy": 0.20503539890050887,
|
|
"num_tokens": 11209656.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.819787073135376,
|
|
"epoch": 5.6961710591745405,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00025907971173207107,
|
|
"loss": 5.2538,
|
|
"mean_token_accuracy": 0.19748033136129378,
|
|
"num_tokens": 11218817.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.96981029510498,
|
|
"epoch": 5.701143709597215,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002586901852603576,
|
|
"loss": 5.3449,
|
|
"mean_token_accuracy": 0.19525426328182222,
|
|
"num_tokens": 11229291.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.904513931274414,
|
|
"epoch": 5.706116360019891,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002583007079237069,
|
|
"loss": 5.3559,
|
|
"mean_token_accuracy": 0.19547263979911805,
|
|
"num_tokens": 11238877.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.913414430618286,
|
|
"epoch": 5.711089010442566,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00025791128089546107,
|
|
"loss": 5.351,
|
|
"mean_token_accuracy": 0.19335026890039445,
|
|
"num_tokens": 11248885.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.92507529258728,
|
|
"epoch": 5.7160616608652415,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00025752190534881065,
|
|
"loss": 5.3162,
|
|
"mean_token_accuracy": 0.1959017127752304,
|
|
"num_tokens": 11259423.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.9086061954498295,
|
|
"epoch": 5.721034311287917,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025713258245679127,
|
|
"loss": 5.3627,
|
|
"mean_token_accuracy": 0.19068257063627242,
|
|
"num_tokens": 11268900.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.849522590637207,
|
|
"epoch": 5.726006961710592,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00025674331339227966,
|
|
"loss": 5.1804,
|
|
"mean_token_accuracy": 0.2123017877340317,
|
|
"num_tokens": 11277997.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.988729619979859,
|
|
"epoch": 5.730979612133267,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00025635409932799074,
|
|
"loss": 5.4611,
|
|
"mean_token_accuracy": 0.1865826889872551,
|
|
"num_tokens": 11288788.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.830811500549316,
|
|
"epoch": 5.7359522625559425,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002559649414364735,
|
|
"loss": 5.2289,
|
|
"mean_token_accuracy": 0.20971091985702514,
|
|
"num_tokens": 11297785.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.779963493347168,
|
|
"epoch": 5.740924912978618,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002555758408901076,
|
|
"loss": 5.2438,
|
|
"mean_token_accuracy": 0.2077435299754143,
|
|
"num_tokens": 11307851.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.897092056274414,
|
|
"epoch": 5.745897563401293,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00025518679886110023,
|
|
"loss": 5.3324,
|
|
"mean_token_accuracy": 0.19966096729040145,
|
|
"num_tokens": 11318616.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.925905990600586,
|
|
"epoch": 5.750870213823968,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002547978165214822,
|
|
"loss": 5.2981,
|
|
"mean_token_accuracy": 0.19964414238929748,
|
|
"num_tokens": 11328751.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.7189911842346195,
|
|
"epoch": 5.7558428642466435,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025440889504310416,
|
|
"loss": 5.0637,
|
|
"mean_token_accuracy": 0.2217484012246132,
|
|
"num_tokens": 11338054.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.898138570785522,
|
|
"epoch": 5.760815514669319,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00025402003559763393,
|
|
"loss": 5.3463,
|
|
"mean_token_accuracy": 0.19602275043725967,
|
|
"num_tokens": 11348341.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.899764776229858,
|
|
"epoch": 5.765788165091994,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025363123935655214,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.19928140789270402,
|
|
"num_tokens": 11357496.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.877427768707276,
|
|
"epoch": 5.770760815514669,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002532425074911491,
|
|
"loss": 5.2731,
|
|
"mean_token_accuracy": 0.2011702299118042,
|
|
"num_tokens": 11366915.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.892586755752563,
|
|
"epoch": 5.7757334659373445,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002528538411725212,
|
|
"loss": 5.3893,
|
|
"mean_token_accuracy": 0.19161875694990158,
|
|
"num_tokens": 11376920.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.746992301940918,
|
|
"epoch": 5.78070611636002,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002524652415715672,
|
|
"loss": 5.1286,
|
|
"mean_token_accuracy": 0.22508858889341354,
|
|
"num_tokens": 11387405.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.962549209594727,
|
|
"epoch": 5.785678766782695,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002520767098589852,
|
|
"loss": 5.453,
|
|
"mean_token_accuracy": 0.1861768737435341,
|
|
"num_tokens": 11396553.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.959508991241455,
|
|
"epoch": 5.79065141720537,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002516882472052683,
|
|
"loss": 5.4015,
|
|
"mean_token_accuracy": 0.19300435483455658,
|
|
"num_tokens": 11406964.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.9503802299499515,
|
|
"epoch": 5.7956240676280455,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002512998547807022,
|
|
"loss": 5.3656,
|
|
"mean_token_accuracy": 0.1968909963965416,
|
|
"num_tokens": 11416550.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.88563494682312,
|
|
"epoch": 5.800596718050721,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025091153375536026,
|
|
"loss": 5.2061,
|
|
"mean_token_accuracy": 0.20241119861602783,
|
|
"num_tokens": 11425477.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.897620820999146,
|
|
"epoch": 5.805569368473396,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025052328529910135,
|
|
"loss": 5.2651,
|
|
"mean_token_accuracy": 0.19973861575126647,
|
|
"num_tokens": 11435153.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.802490091323852,
|
|
"epoch": 5.810542018896071,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002501351105815655,
|
|
"loss": 5.2419,
|
|
"mean_token_accuracy": 0.20793721526861192,
|
|
"num_tokens": 11445635.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.9307701587677,
|
|
"epoch": 5.8155146693187465,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024974701077217057,
|
|
"loss": 5.4315,
|
|
"mean_token_accuracy": 0.18427260369062423,
|
|
"num_tokens": 11455526.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.913471317291259,
|
|
"epoch": 5.820487319741423,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024935898704010875,
|
|
"loss": 5.281,
|
|
"mean_token_accuracy": 0.19725342392921447,
|
|
"num_tokens": 11464115.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.764231491088867,
|
|
"epoch": 5.825459970164097,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00024897104055434304,
|
|
"loss": 5.1812,
|
|
"mean_token_accuracy": 0.21481697857379914,
|
|
"num_tokens": 11475456.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.7597497463226315,
|
|
"epoch": 5.830432620586773,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024858317248360387,
|
|
"loss": 5.1059,
|
|
"mean_token_accuracy": 0.22116924077272415,
|
|
"num_tokens": 11485796.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.95292854309082,
|
|
"epoch": 5.835405271009448,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002481953839963851,
|
|
"loss": 5.3525,
|
|
"mean_token_accuracy": 0.18911842852830887,
|
|
"num_tokens": 11494904.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.740696620941162,
|
|
"epoch": 5.840377921432124,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002478076762609413,
|
|
"loss": 5.1233,
|
|
"mean_token_accuracy": 0.2223310500383377,
|
|
"num_tokens": 11504429.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.90019154548645,
|
|
"epoch": 5.845350571854799,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002474200504452834,
|
|
"loss": 5.3299,
|
|
"mean_token_accuracy": 0.19304918646812438,
|
|
"num_tokens": 11515557.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.960551309585571,
|
|
"epoch": 5.850323222277474,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024703250771717576,
|
|
"loss": 5.3931,
|
|
"mean_token_accuracy": 0.19364458173513413,
|
|
"num_tokens": 11525613.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.938970136642456,
|
|
"epoch": 5.855295872700149,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024664504924413224,
|
|
"loss": 5.3171,
|
|
"mean_token_accuracy": 0.19158960580825807,
|
|
"num_tokens": 11535463.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.899830722808838,
|
|
"epoch": 5.860268523122825,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002462576761934129,
|
|
"loss": 5.2805,
|
|
"mean_token_accuracy": 0.20016167014837266,
|
|
"num_tokens": 11546277.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.905777502059936,
|
|
"epoch": 5.8652411735455,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002458703897320208,
|
|
"loss": 5.3219,
|
|
"mean_token_accuracy": 0.19434359669685364,
|
|
"num_tokens": 11555361.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.881389904022217,
|
|
"epoch": 5.870213823968175,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024548319102669745,
|
|
"loss": 5.2067,
|
|
"mean_token_accuracy": 0.19962632656097412,
|
|
"num_tokens": 11564976.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.801442241668701,
|
|
"epoch": 5.87518647439085,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024509608124392065,
|
|
"loss": 5.1732,
|
|
"mean_token_accuracy": 0.20508586764335632,
|
|
"num_tokens": 11574574.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.87659912109375,
|
|
"epoch": 5.880159124813526,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002447090615499,
|
|
"loss": 5.3123,
|
|
"mean_token_accuracy": 0.19213323444128036,
|
|
"num_tokens": 11584571.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.968310356140137,
|
|
"epoch": 5.885131775236201,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024432213311057383,
|
|
"loss": 5.4106,
|
|
"mean_token_accuracy": 0.18745146691799164,
|
|
"num_tokens": 11594829.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.919745588302613,
|
|
"epoch": 5.890104425658876,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024393529709160544,
|
|
"loss": 5.368,
|
|
"mean_token_accuracy": 0.19908616542816163,
|
|
"num_tokens": 11604520.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.966621732711792,
|
|
"epoch": 5.895077076081551,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024354855465837967,
|
|
"loss": 5.2728,
|
|
"mean_token_accuracy": 0.19953136891126633,
|
|
"num_tokens": 11612909.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.936744594573975,
|
|
"epoch": 5.900049726504227,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024316190697599955,
|
|
"loss": 5.4184,
|
|
"mean_token_accuracy": 0.18667218387126921,
|
|
"num_tokens": 11622928.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.934949159622192,
|
|
"epoch": 5.905022376926902,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024277535520928257,
|
|
"loss": 5.4365,
|
|
"mean_token_accuracy": 0.18724340051412583,
|
|
"num_tokens": 11632192.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.88970251083374,
|
|
"epoch": 5.909995027349577,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024238890052275742,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.2046724319458008,
|
|
"num_tokens": 11642346.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.933574056625366,
|
|
"epoch": 5.914967677772252,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024200254408066014,
|
|
"loss": 5.248,
|
|
"mean_token_accuracy": 0.20361822843551636,
|
|
"num_tokens": 11651527.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.936920261383056,
|
|
"epoch": 5.919940328194928,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002416162870469308,
|
|
"loss": 5.3478,
|
|
"mean_token_accuracy": 0.19046850353479386,
|
|
"num_tokens": 11660920.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.8736261367797855,
|
|
"epoch": 5.924912978617603,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002412301305852102,
|
|
"loss": 5.317,
|
|
"mean_token_accuracy": 0.1956012114882469,
|
|
"num_tokens": 11670220.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.739864921569824,
|
|
"epoch": 5.929885629040278,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024084407585883588,
|
|
"loss": 5.1189,
|
|
"mean_token_accuracy": 0.22900400608778,
|
|
"num_tokens": 11680922.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.917207431793213,
|
|
"epoch": 5.934858279462953,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002404581240308391,
|
|
"loss": 5.2927,
|
|
"mean_token_accuracy": 0.20121543258428573,
|
|
"num_tokens": 11691138.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.918303394317627,
|
|
"epoch": 5.939830929885629,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000240072276263941,
|
|
"loss": 5.335,
|
|
"mean_token_accuracy": 0.19545186460018157,
|
|
"num_tokens": 11700379.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 6.008311414718628,
|
|
"epoch": 5.944803580308305,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00023968653372054933,
|
|
"loss": 5.3363,
|
|
"mean_token_accuracy": 0.1929083913564682,
|
|
"num_tokens": 11709678.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.841628789901733,
|
|
"epoch": 5.949776230730979,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00023930089756275474,
|
|
"loss": 5.1339,
|
|
"mean_token_accuracy": 0.2087523505091667,
|
|
"num_tokens": 11719700.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.870734453201294,
|
|
"epoch": 5.954748881153655,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002389153689523275,
|
|
"loss": 5.3072,
|
|
"mean_token_accuracy": 0.1919586405158043,
|
|
"num_tokens": 11730087.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.893419694900513,
|
|
"epoch": 5.9597215315763306,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023852994905071385,
|
|
"loss": 5.3225,
|
|
"mean_token_accuracy": 0.19409748464822768,
|
|
"num_tokens": 11740063.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 6.0444221019744875,
|
|
"epoch": 5.964694181999006,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023814463901903238,
|
|
"loss": 5.4121,
|
|
"mean_token_accuracy": 0.1901340514421463,
|
|
"num_tokens": 11750213.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.964694181999006,
|
|
"eval_entropy": 5.877788833758798,
|
|
"eval_loss": 6.512768268585205,
|
|
"eval_mean_token_accuracy": 0.14642375173725292,
|
|
"eval_num_tokens": 11750213.0,
|
|
"eval_runtime": 11.2789,
|
|
"eval_samples_per_second": 307.21,
|
|
"eval_steps_per_second": 38.479,
|
|
"step": 6000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10050,
|
|
"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": 1.9250696581632e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|