1035 lines
27 KiB
JSON
1035 lines
27 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.011532163203173651,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691907501220703,
|
|
"epoch": 0.0001153216320317365,
|
|
"grad_norm": 17.375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.5245,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 3965.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691843128204345,
|
|
"epoch": 0.000230643264063473,
|
|
"grad_norm": 18.5,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.3993,
|
|
"mean_token_accuracy": 0.0003649635007604957,
|
|
"num_tokens": 7510.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.69088363647461,
|
|
"epoch": 0.0003459648960952095,
|
|
"grad_norm": 13.125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.1564,
|
|
"mean_token_accuracy": 0.03439897168427706,
|
|
"num_tokens": 11084.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.681573295593262,
|
|
"epoch": 0.000461286528126946,
|
|
"grad_norm": 9.0625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 9.7517,
|
|
"mean_token_accuracy": 0.06451544389128686,
|
|
"num_tokens": 14319.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.623183345794677,
|
|
"epoch": 0.0005766081601586826,
|
|
"grad_norm": 5.8125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.4401,
|
|
"mean_token_accuracy": 0.06273870654404164,
|
|
"num_tokens": 17670.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.555589962005616,
|
|
"epoch": 0.000691929792190419,
|
|
"grad_norm": 4.9375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.2749,
|
|
"mean_token_accuracy": 0.05860213525593281,
|
|
"num_tokens": 21177.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.556735229492187,
|
|
"epoch": 0.0008072514242221556,
|
|
"grad_norm": 4.25,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.1316,
|
|
"mean_token_accuracy": 0.06961804144084453,
|
|
"num_tokens": 24417.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.520895004272461,
|
|
"epoch": 0.000922573056253892,
|
|
"grad_norm": 4.09375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.1409,
|
|
"mean_token_accuracy": 0.067233781889081,
|
|
"num_tokens": 27824.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.48238296508789,
|
|
"epoch": 0.0010378946882856287,
|
|
"grad_norm": 3.828125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.1384,
|
|
"mean_token_accuracy": 0.06338647790253163,
|
|
"num_tokens": 31712.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.46641788482666,
|
|
"epoch": 0.0011532163203173652,
|
|
"grad_norm": 3.640625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.0513,
|
|
"mean_token_accuracy": 0.07285529412329197,
|
|
"num_tokens": 35207.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.364572525024414,
|
|
"epoch": 0.0012685379523491016,
|
|
"grad_norm": 3.28125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.0362,
|
|
"mean_token_accuracy": 0.06746394783258439,
|
|
"num_tokens": 39066.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.27311315536499,
|
|
"epoch": 0.001383859584380838,
|
|
"grad_norm": 3.5,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 8.8148,
|
|
"mean_token_accuracy": 0.07908576317131519,
|
|
"num_tokens": 42389.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.151123905181885,
|
|
"epoch": 0.0014991812164125747,
|
|
"grad_norm": 2.953125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 8.8459,
|
|
"mean_token_accuracy": 0.07942395582795143,
|
|
"num_tokens": 45778.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.179048538208008,
|
|
"epoch": 0.0016145028484443112,
|
|
"grad_norm": 3.0625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 8.6928,
|
|
"mean_token_accuracy": 0.07802934609353543,
|
|
"num_tokens": 48918.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.147464656829834,
|
|
"epoch": 0.0017298244804760477,
|
|
"grad_norm": 4.0625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 8.6836,
|
|
"mean_token_accuracy": 0.07859932407736778,
|
|
"num_tokens": 52456.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.119709587097168,
|
|
"epoch": 0.001845146112507784,
|
|
"grad_norm": 3.046875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.6315,
|
|
"mean_token_accuracy": 0.07672536484897137,
|
|
"num_tokens": 56193.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.134781169891358,
|
|
"epoch": 0.0019604677445395208,
|
|
"grad_norm": 3.5,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.5835,
|
|
"mean_token_accuracy": 0.07438027523458005,
|
|
"num_tokens": 59977.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.018630409240723,
|
|
"epoch": 0.0020757893765712574,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.5291,
|
|
"mean_token_accuracy": 0.0768112525343895,
|
|
"num_tokens": 63844.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 9.969991493225098,
|
|
"epoch": 0.0021911110086029937,
|
|
"grad_norm": 2.875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.4302,
|
|
"mean_token_accuracy": 0.07833829969167709,
|
|
"num_tokens": 67356.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.947973537445069,
|
|
"epoch": 0.0023064326406347303,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.2975,
|
|
"mean_token_accuracy": 0.08131772689521313,
|
|
"num_tokens": 70767.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.674797916412354,
|
|
"epoch": 0.0024217542726664666,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.2983,
|
|
"mean_token_accuracy": 0.08184195645153522,
|
|
"num_tokens": 74437.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.848657512664795,
|
|
"epoch": 0.0025370759046982033,
|
|
"grad_norm": 2.96875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.2258,
|
|
"mean_token_accuracy": 0.07810623683035374,
|
|
"num_tokens": 77940.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.696725845336914,
|
|
"epoch": 0.00265239753672994,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.1357,
|
|
"mean_token_accuracy": 0.07201453074812889,
|
|
"num_tokens": 81669.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.512047672271729,
|
|
"epoch": 0.002767719168761676,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.0767,
|
|
"mean_token_accuracy": 0.0643788930028677,
|
|
"num_tokens": 86135.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.434719371795655,
|
|
"epoch": 0.002883040800793413,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.898,
|
|
"mean_token_accuracy": 0.0802077766507864,
|
|
"num_tokens": 89620.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.149808502197265,
|
|
"epoch": 0.0029983624328251495,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.7739,
|
|
"mean_token_accuracy": 0.08522589541971684,
|
|
"num_tokens": 93245.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.161969184875488,
|
|
"epoch": 0.0031136840648568857,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.6885,
|
|
"mean_token_accuracy": 0.09521296098828316,
|
|
"num_tokens": 96754.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.86840353012085,
|
|
"epoch": 0.0032290056968886224,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6545,
|
|
"mean_token_accuracy": 0.08669420927762986,
|
|
"num_tokens": 100266.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.901450920104981,
|
|
"epoch": 0.003344327328920359,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5653,
|
|
"mean_token_accuracy": 0.07821874283254146,
|
|
"num_tokens": 103824.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.649330520629883,
|
|
"epoch": 0.0034596489609520953,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5483,
|
|
"mean_token_accuracy": 0.0837211973965168,
|
|
"num_tokens": 107659.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.520522785186767,
|
|
"epoch": 0.003574970592983832,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3799,
|
|
"mean_token_accuracy": 0.08768085911870002,
|
|
"num_tokens": 111348.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.380702209472656,
|
|
"epoch": 0.003690292225015568,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4222,
|
|
"mean_token_accuracy": 0.08212160468101501,
|
|
"num_tokens": 115386.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.227167320251464,
|
|
"epoch": 0.003805613857047305,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.3087,
|
|
"mean_token_accuracy": 0.09835789948701859,
|
|
"num_tokens": 118772.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.145322608947755,
|
|
"epoch": 0.0039209354890790415,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.416,
|
|
"mean_token_accuracy": 0.09204246997833251,
|
|
"num_tokens": 122680.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.233028316497803,
|
|
"epoch": 0.004036257121110778,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.3521,
|
|
"mean_token_accuracy": 0.08081735149025918,
|
|
"num_tokens": 127162.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.950575256347657,
|
|
"epoch": 0.004151578753142515,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2055,
|
|
"mean_token_accuracy": 0.09383777529001236,
|
|
"num_tokens": 130684.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.000711679458618,
|
|
"epoch": 0.004266900385174251,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.275,
|
|
"mean_token_accuracy": 0.09546211659908295,
|
|
"num_tokens": 134154.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.941235446929932,
|
|
"epoch": 0.004382222017205987,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2944,
|
|
"mean_token_accuracy": 0.08537636548280716,
|
|
"num_tokens": 137811.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.906629467010498,
|
|
"epoch": 0.004497543649237724,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.1824,
|
|
"mean_token_accuracy": 0.09572790712118148,
|
|
"num_tokens": 141220.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.942177057266235,
|
|
"epoch": 0.004612865281269461,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2036,
|
|
"mean_token_accuracy": 0.09937367662787437,
|
|
"num_tokens": 144768.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.685758304595947,
|
|
"epoch": 0.004728186913301197,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.137,
|
|
"mean_token_accuracy": 0.09695517122745514,
|
|
"num_tokens": 148173.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.714945936203003,
|
|
"epoch": 0.004843508545332933,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1256,
|
|
"mean_token_accuracy": 0.10458688139915466,
|
|
"num_tokens": 151861.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.71102352142334,
|
|
"epoch": 0.00495883017736467,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.066,
|
|
"mean_token_accuracy": 0.10715894550085067,
|
|
"num_tokens": 155191.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.614436388015747,
|
|
"epoch": 0.0050741518093964065,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1459,
|
|
"mean_token_accuracy": 0.09644531235098838,
|
|
"num_tokens": 158655.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.692019128799439,
|
|
"epoch": 0.005189473441428143,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1231,
|
|
"mean_token_accuracy": 0.10170018300414085,
|
|
"num_tokens": 162244.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.713358163833618,
|
|
"epoch": 0.00530479507345988,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2913,
|
|
"mean_token_accuracy": 0.08797733411192894,
|
|
"num_tokens": 165887.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.608382606506348,
|
|
"epoch": 0.0054201167054916165,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.0288,
|
|
"mean_token_accuracy": 0.10224084109067917,
|
|
"num_tokens": 169248.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.681228685379028,
|
|
"epoch": 0.005535438337523352,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.137,
|
|
"mean_token_accuracy": 0.09048556201159955,
|
|
"num_tokens": 172559.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.473513507843018,
|
|
"epoch": 0.005650759969555089,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0968,
|
|
"mean_token_accuracy": 0.1024228110909462,
|
|
"num_tokens": 175961.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.584992599487305,
|
|
"epoch": 0.005766081601586826,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1516,
|
|
"mean_token_accuracy": 0.09878315702080727,
|
|
"num_tokens": 179611.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.685141277313233,
|
|
"epoch": 0.005881403233618562,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1129,
|
|
"mean_token_accuracy": 0.10327758342027664,
|
|
"num_tokens": 183105.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.646425008773804,
|
|
"epoch": 0.005996724865650299,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.118,
|
|
"mean_token_accuracy": 0.09364504367113113,
|
|
"num_tokens": 186682.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.5202278137207035,
|
|
"epoch": 0.006112046497682035,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1542,
|
|
"mean_token_accuracy": 0.09994478896260262,
|
|
"num_tokens": 190487.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.5590576171875,
|
|
"epoch": 0.0062273681297137715,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1467,
|
|
"mean_token_accuracy": 0.0977135255932808,
|
|
"num_tokens": 194445.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.514654350280762,
|
|
"epoch": 0.006342689761745508,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0653,
|
|
"mean_token_accuracy": 0.10815930888056755,
|
|
"num_tokens": 197864.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.6202771186828615,
|
|
"epoch": 0.006458011393777245,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.99,
|
|
"mean_token_accuracy": 0.1028586745262146,
|
|
"num_tokens": 201465.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.317496013641358,
|
|
"epoch": 0.0065733330258089815,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9834,
|
|
"mean_token_accuracy": 0.09826496616005898,
|
|
"num_tokens": 204918.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.434334373474121,
|
|
"epoch": 0.006688654657840718,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.078,
|
|
"mean_token_accuracy": 0.10162978768348693,
|
|
"num_tokens": 208725.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.4441381931304935,
|
|
"epoch": 0.006803976289872454,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0367,
|
|
"mean_token_accuracy": 0.10961202383041382,
|
|
"num_tokens": 212286.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.442938041687012,
|
|
"epoch": 0.006919297921904191,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9635,
|
|
"mean_token_accuracy": 0.11073270812630653,
|
|
"num_tokens": 215900.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.303906488418579,
|
|
"epoch": 0.007034619553935927,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9921,
|
|
"mean_token_accuracy": 0.10271828547120095,
|
|
"num_tokens": 219495.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.581103324890137,
|
|
"epoch": 0.007149941185967664,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0565,
|
|
"mean_token_accuracy": 0.09874569922685623,
|
|
"num_tokens": 223358.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.358316278457641,
|
|
"epoch": 0.007265262817999401,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9374,
|
|
"mean_token_accuracy": 0.11364061161875724,
|
|
"num_tokens": 226549.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.333794832229614,
|
|
"epoch": 0.007380584450031136,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0438,
|
|
"mean_token_accuracy": 0.11026700586080551,
|
|
"num_tokens": 230102.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.446019983291626,
|
|
"epoch": 0.007495906082062873,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9717,
|
|
"mean_token_accuracy": 0.10468028411269188,
|
|
"num_tokens": 233652.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.4383687496185305,
|
|
"epoch": 0.00761122771409461,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9563,
|
|
"mean_token_accuracy": 0.10809829756617546,
|
|
"num_tokens": 237100.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.277336835861206,
|
|
"epoch": 0.007726549346126346,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.835,
|
|
"mean_token_accuracy": 0.1123481884598732,
|
|
"num_tokens": 240390.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.367178535461425,
|
|
"epoch": 0.007841870978158083,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9353,
|
|
"mean_token_accuracy": 0.10941664353013039,
|
|
"num_tokens": 243955.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.226317501068115,
|
|
"epoch": 0.007957192610189819,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8903,
|
|
"mean_token_accuracy": 0.11319939866662025,
|
|
"num_tokens": 247371.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.388652610778808,
|
|
"epoch": 0.008072514242221556,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9579,
|
|
"mean_token_accuracy": 0.10422090664505959,
|
|
"num_tokens": 250794.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.3202920913696286,
|
|
"epoch": 0.008187835874253292,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.985,
|
|
"mean_token_accuracy": 0.10096767656505108,
|
|
"num_tokens": 254434.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.292840766906738,
|
|
"epoch": 0.00830315750628503,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9804,
|
|
"mean_token_accuracy": 0.10753846392035485,
|
|
"num_tokens": 258039.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.453191423416138,
|
|
"epoch": 0.008418479138316766,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9874,
|
|
"mean_token_accuracy": 0.11508287489414215,
|
|
"num_tokens": 261851.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.177758312225341,
|
|
"epoch": 0.008533800770348501,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9374,
|
|
"mean_token_accuracy": 0.10858769118785858,
|
|
"num_tokens": 265903.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.471200084686279,
|
|
"epoch": 0.008649122402380239,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.9962,
|
|
"mean_token_accuracy": 0.10989872291684151,
|
|
"num_tokens": 269471.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.166895580291748,
|
|
"epoch": 0.008764444034411975,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9137,
|
|
"mean_token_accuracy": 0.11640691384673119,
|
|
"num_tokens": 272945.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.404301357269287,
|
|
"epoch": 0.008879765666443712,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9117,
|
|
"mean_token_accuracy": 0.1151728942990303,
|
|
"num_tokens": 276300.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.1740296363830565,
|
|
"epoch": 0.008995087298475448,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9439,
|
|
"mean_token_accuracy": 0.10883786007761956,
|
|
"num_tokens": 279930.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.307166481018067,
|
|
"epoch": 0.009110408930507184,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.92,
|
|
"mean_token_accuracy": 0.1120470218360424,
|
|
"num_tokens": 283709.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.349106884002685,
|
|
"epoch": 0.009225730562538921,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0302,
|
|
"mean_token_accuracy": 0.11455085650086402,
|
|
"num_tokens": 287880.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.221231746673584,
|
|
"epoch": 0.009341052194570657,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8002,
|
|
"mean_token_accuracy": 0.11782657876610755,
|
|
"num_tokens": 291367.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.293564319610596,
|
|
"epoch": 0.009456373826602395,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.9735,
|
|
"mean_token_accuracy": 0.10948061645030975,
|
|
"num_tokens": 294920.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.203423309326172,
|
|
"epoch": 0.00957169545863413,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8951,
|
|
"mean_token_accuracy": 0.1170351468026638,
|
|
"num_tokens": 298347.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.1942479610443115,
|
|
"epoch": 0.009687017090665866,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7875,
|
|
"mean_token_accuracy": 0.1238088421523571,
|
|
"num_tokens": 301787.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.0459576606750485,
|
|
"epoch": 0.009802338722697604,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8287,
|
|
"mean_token_accuracy": 0.12091329768300056,
|
|
"num_tokens": 305284.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.336889219284058,
|
|
"epoch": 0.00991766035472934,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.9463,
|
|
"mean_token_accuracy": 0.11388053148984909,
|
|
"num_tokens": 308963.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.106842947006226,
|
|
"epoch": 0.010032981986761077,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.809,
|
|
"mean_token_accuracy": 0.12353090792894364,
|
|
"num_tokens": 312422.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.182272291183471,
|
|
"epoch": 0.010148303618792813,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7662,
|
|
"mean_token_accuracy": 0.11858817338943481,
|
|
"num_tokens": 316194.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.024605417251587,
|
|
"epoch": 0.010263625250824549,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8614,
|
|
"mean_token_accuracy": 0.11561542153358459,
|
|
"num_tokens": 319824.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.047331666946411,
|
|
"epoch": 0.010378946882856286,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7707,
|
|
"mean_token_accuracy": 0.1250594787299633,
|
|
"num_tokens": 323503.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.204434967041015,
|
|
"epoch": 0.010494268514888022,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8941,
|
|
"mean_token_accuracy": 0.11519786417484283,
|
|
"num_tokens": 326858.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.106210565567016,
|
|
"epoch": 0.01060959014691976,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8793,
|
|
"mean_token_accuracy": 0.11701082810759544,
|
|
"num_tokens": 330370.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.177642726898194,
|
|
"epoch": 0.010724911778951495,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.11765560358762742,
|
|
"num_tokens": 334041.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.11465482711792,
|
|
"epoch": 0.010840233410983233,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7651,
|
|
"mean_token_accuracy": 0.12222857922315597,
|
|
"num_tokens": 337724.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.927467727661133,
|
|
"epoch": 0.010955555043014969,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6746,
|
|
"mean_token_accuracy": 0.12812104001641272,
|
|
"num_tokens": 340946.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.083363389968872,
|
|
"epoch": 0.011070876675046705,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8418,
|
|
"mean_token_accuracy": 0.12921102717518806,
|
|
"num_tokens": 344603.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.131814002990723,
|
|
"epoch": 0.011186198307078442,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8164,
|
|
"mean_token_accuracy": 0.11969970613718033,
|
|
"num_tokens": 348153.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.9778149127960205,
|
|
"epoch": 0.011301519939110178,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8872,
|
|
"mean_token_accuracy": 0.12184126004576683,
|
|
"num_tokens": 351682.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.12193341255188,
|
|
"epoch": 0.011416841571141916,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7129,
|
|
"mean_token_accuracy": 0.12792302519083024,
|
|
"num_tokens": 355245.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.980188322067261,
|
|
"epoch": 0.011532163203173651,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8127,
|
|
"mean_token_accuracy": 0.12161365896463394,
|
|
"num_tokens": 358888.0,
|
|
"step": 500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"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": 551334352896000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|