5090 lines
138 KiB
JSON
5090 lines
138 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.485330681253108,
|
|
"eval_steps": 500,
|
|
"global_step": 2500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 8057037791232000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|