8035 lines
219 KiB
JSON
8035 lines
219 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.09225730562538921,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692046642303467,
|
|
"epoch": 0.0001153216320317365,
|
|
"grad_norm": 16.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.4811,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 3293.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.692021083831786,
|
|
"epoch": 0.000230643264063473,
|
|
"grad_norm": 18.875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.4344,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 6743.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691608142852782,
|
|
"epoch": 0.0003459648960952095,
|
|
"grad_norm": 13.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.184,
|
|
"mean_token_accuracy": 0.01596597593743354,
|
|
"num_tokens": 10261.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.686354446411134,
|
|
"epoch": 0.000461286528126946,
|
|
"grad_norm": 9.4375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 9.8654,
|
|
"mean_token_accuracy": 0.04747712537646294,
|
|
"num_tokens": 14153.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.65447473526001,
|
|
"epoch": 0.0005766081601586826,
|
|
"grad_norm": 6.3125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.5581,
|
|
"mean_token_accuracy": 0.0560380831360817,
|
|
"num_tokens": 17774.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.609045124053955,
|
|
"epoch": 0.000691929792190419,
|
|
"grad_norm": 4.75,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.3551,
|
|
"mean_token_accuracy": 0.06091434173285961,
|
|
"num_tokens": 21521.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.577239322662354,
|
|
"epoch": 0.0008072514242221556,
|
|
"grad_norm": 4.75,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.2179,
|
|
"mean_token_accuracy": 0.06620409153401852,
|
|
"num_tokens": 24880.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.546881198883057,
|
|
"epoch": 0.000922573056253892,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.1091,
|
|
"mean_token_accuracy": 0.06204554140567779,
|
|
"num_tokens": 28167.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.515935897827148,
|
|
"epoch": 0.0010378946882856287,
|
|
"grad_norm": 3.40625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.1072,
|
|
"mean_token_accuracy": 0.06435491405427456,
|
|
"num_tokens": 31745.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.450439167022704,
|
|
"epoch": 0.0011532163203173652,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.0469,
|
|
"mean_token_accuracy": 0.06713954657316208,
|
|
"num_tokens": 35270.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.369979953765869,
|
|
"epoch": 0.0012685379523491016,
|
|
"grad_norm": 3.328125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.0259,
|
|
"mean_token_accuracy": 0.07268906272947788,
|
|
"num_tokens": 38943.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.2349365234375,
|
|
"epoch": 0.001383859584380838,
|
|
"grad_norm": 4.1875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 8.8073,
|
|
"mean_token_accuracy": 0.07592999711632728,
|
|
"num_tokens": 42299.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.174158954620362,
|
|
"epoch": 0.0014991812164125747,
|
|
"grad_norm": 2.984375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 8.8684,
|
|
"mean_token_accuracy": 0.07179933935403823,
|
|
"num_tokens": 45661.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.246811008453369,
|
|
"epoch": 0.0016145028484443112,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 8.951,
|
|
"mean_token_accuracy": 0.061866828426718715,
|
|
"num_tokens": 49683.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.264655303955077,
|
|
"epoch": 0.0017298244804760477,
|
|
"grad_norm": 2.921875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 8.716,
|
|
"mean_token_accuracy": 0.07860686406493186,
|
|
"num_tokens": 52752.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.173390865325928,
|
|
"epoch": 0.001845146112507784,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.6778,
|
|
"mean_token_accuracy": 0.07509532310068608,
|
|
"num_tokens": 56367.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.120823097229003,
|
|
"epoch": 0.0019604677445395208,
|
|
"grad_norm": 3.46875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.596,
|
|
"mean_token_accuracy": 0.08393591865897179,
|
|
"num_tokens": 59967.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.119685459136964,
|
|
"epoch": 0.0020757893765712574,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.5686,
|
|
"mean_token_accuracy": 0.07523884475231171,
|
|
"num_tokens": 63634.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.038768196105957,
|
|
"epoch": 0.0021911110086029937,
|
|
"grad_norm": 3.015625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.3464,
|
|
"mean_token_accuracy": 0.0838980995118618,
|
|
"num_tokens": 66883.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.92626495361328,
|
|
"epoch": 0.0023064326406347303,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.3612,
|
|
"mean_token_accuracy": 0.0789190474897623,
|
|
"num_tokens": 70491.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.932885265350341,
|
|
"epoch": 0.0024217542726664666,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.2115,
|
|
"mean_token_accuracy": 0.08061706572771073,
|
|
"num_tokens": 73852.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.765224361419678,
|
|
"epoch": 0.0025370759046982033,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.164,
|
|
"mean_token_accuracy": 0.0914115659892559,
|
|
"num_tokens": 77442.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.612014007568359,
|
|
"epoch": 0.00265239753672994,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.0555,
|
|
"mean_token_accuracy": 0.08430748134851455,
|
|
"num_tokens": 80871.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.554380130767822,
|
|
"epoch": 0.002767719168761676,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.9495,
|
|
"mean_token_accuracy": 0.09184777326881885,
|
|
"num_tokens": 84444.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.397439956665039,
|
|
"epoch": 0.002883040800793413,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.9236,
|
|
"mean_token_accuracy": 0.0814009454101324,
|
|
"num_tokens": 87939.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.200541114807129,
|
|
"epoch": 0.0029983624328251495,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.6689,
|
|
"mean_token_accuracy": 0.08414206653833389,
|
|
"num_tokens": 91189.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.055619144439698,
|
|
"epoch": 0.0031136840648568857,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.6462,
|
|
"mean_token_accuracy": 0.08812693618237973,
|
|
"num_tokens": 94777.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.0249493598938,
|
|
"epoch": 0.0032290056968886224,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6653,
|
|
"mean_token_accuracy": 0.08522222451865673,
|
|
"num_tokens": 98581.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.796249675750733,
|
|
"epoch": 0.003344327328920359,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6153,
|
|
"mean_token_accuracy": 0.08283311799168587,
|
|
"num_tokens": 102333.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.677543926239014,
|
|
"epoch": 0.0034596489609520953,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5304,
|
|
"mean_token_accuracy": 0.08452097214758396,
|
|
"num_tokens": 106267.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.623601245880128,
|
|
"epoch": 0.003574970592983832,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.4307,
|
|
"mean_token_accuracy": 0.08808897696435451,
|
|
"num_tokens": 109989.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.373943901062011,
|
|
"epoch": 0.003690292225015568,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4063,
|
|
"mean_token_accuracy": 0.08707165420055389,
|
|
"num_tokens": 113378.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.395255756378173,
|
|
"epoch": 0.003805613857047305,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.3032,
|
|
"mean_token_accuracy": 0.09222758859395981,
|
|
"num_tokens": 116736.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.167229080200196,
|
|
"epoch": 0.0039209354890790415,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2397,
|
|
"mean_token_accuracy": 0.08980661779642105,
|
|
"num_tokens": 120369.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.110415983200074,
|
|
"epoch": 0.004036257121110778,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.2388,
|
|
"mean_token_accuracy": 0.10482370853424072,
|
|
"num_tokens": 124003.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.005911922454834,
|
|
"epoch": 0.004151578753142515,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2327,
|
|
"mean_token_accuracy": 0.09916695356369018,
|
|
"num_tokens": 127611.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.034901094436645,
|
|
"epoch": 0.004266900385174251,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.255,
|
|
"mean_token_accuracy": 0.09181497618556023,
|
|
"num_tokens": 130837.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.924763011932373,
|
|
"epoch": 0.004382222017205987,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2781,
|
|
"mean_token_accuracy": 0.09120299257338046,
|
|
"num_tokens": 134801.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.028991937637329,
|
|
"epoch": 0.004497543649237724,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3143,
|
|
"mean_token_accuracy": 0.09299080073833466,
|
|
"num_tokens": 138479.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.826745653152466,
|
|
"epoch": 0.004612865281269461,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2351,
|
|
"mean_token_accuracy": 0.09592275992035866,
|
|
"num_tokens": 142536.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.951525592803955,
|
|
"epoch": 0.004728186913301197,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2783,
|
|
"mean_token_accuracy": 0.09658894389867782,
|
|
"num_tokens": 146633.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.681295728683471,
|
|
"epoch": 0.004843508545332933,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1036,
|
|
"mean_token_accuracy": 0.09946461021900177,
|
|
"num_tokens": 150070.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.870249176025391,
|
|
"epoch": 0.00495883017736467,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2654,
|
|
"mean_token_accuracy": 0.08741186410188675,
|
|
"num_tokens": 153698.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.710623836517334,
|
|
"epoch": 0.0050741518093964065,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1792,
|
|
"mean_token_accuracy": 0.09874420538544655,
|
|
"num_tokens": 157381.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.73014268875122,
|
|
"epoch": 0.005189473441428143,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1647,
|
|
"mean_token_accuracy": 0.09424602091312409,
|
|
"num_tokens": 160862.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.8334917545318605,
|
|
"epoch": 0.00530479507345988,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1663,
|
|
"mean_token_accuracy": 0.0998285748064518,
|
|
"num_tokens": 164231.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.634814500808716,
|
|
"epoch": 0.0054201167054916165,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2333,
|
|
"mean_token_accuracy": 0.08781378865242004,
|
|
"num_tokens": 168351.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.6461118221282955,
|
|
"epoch": 0.005535438337523352,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0404,
|
|
"mean_token_accuracy": 0.10981863886117935,
|
|
"num_tokens": 171645.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.518331480026245,
|
|
"epoch": 0.005650759969555089,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0655,
|
|
"mean_token_accuracy": 0.10514177232980729,
|
|
"num_tokens": 175279.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.588093042373657,
|
|
"epoch": 0.005766081601586826,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1508,
|
|
"mean_token_accuracy": 0.0997185155749321,
|
|
"num_tokens": 179111.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.750802040100098,
|
|
"epoch": 0.005881403233618562,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1045,
|
|
"mean_token_accuracy": 0.10787053629755974,
|
|
"num_tokens": 182369.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.515914821624756,
|
|
"epoch": 0.005996724865650299,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0753,
|
|
"mean_token_accuracy": 0.11265580952167512,
|
|
"num_tokens": 185671.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.501624631881714,
|
|
"epoch": 0.006112046497682035,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9932,
|
|
"mean_token_accuracy": 0.1023002490401268,
|
|
"num_tokens": 189295.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.487598466873169,
|
|
"epoch": 0.0062273681297137715,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.0035,
|
|
"mean_token_accuracy": 0.10346534922719001,
|
|
"num_tokens": 192791.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.491527700424195,
|
|
"epoch": 0.006342689761745508,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0257,
|
|
"mean_token_accuracy": 0.10039645433425903,
|
|
"num_tokens": 196550.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.525121784210205,
|
|
"epoch": 0.006458011393777245,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0112,
|
|
"mean_token_accuracy": 0.10743131414055825,
|
|
"num_tokens": 200274.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.3943414211273195,
|
|
"epoch": 0.0065733330258089815,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9813,
|
|
"mean_token_accuracy": 0.11192047148942948,
|
|
"num_tokens": 203748.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.610325717926026,
|
|
"epoch": 0.006688654657840718,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9901,
|
|
"mean_token_accuracy": 0.10699754059314728,
|
|
"num_tokens": 207203.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.490588426589966,
|
|
"epoch": 0.006803976289872454,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1777,
|
|
"mean_token_accuracy": 0.09906014949083328,
|
|
"num_tokens": 210761.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.500645971298217,
|
|
"epoch": 0.006919297921904191,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0411,
|
|
"mean_token_accuracy": 0.11081741005182266,
|
|
"num_tokens": 214362.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.4556807518005375,
|
|
"epoch": 0.007034619553935927,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0964,
|
|
"mean_token_accuracy": 0.10465478524565697,
|
|
"num_tokens": 217938.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.400477266311645,
|
|
"epoch": 0.007149941185967664,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0187,
|
|
"mean_token_accuracy": 0.102249476313591,
|
|
"num_tokens": 221519.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.4884284973144535,
|
|
"epoch": 0.007265262817999401,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9382,
|
|
"mean_token_accuracy": 0.11059758365154267,
|
|
"num_tokens": 224918.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.453373765945434,
|
|
"epoch": 0.007380584450031136,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.97,
|
|
"mean_token_accuracy": 0.11169943362474441,
|
|
"num_tokens": 228334.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.396696710586548,
|
|
"epoch": 0.007495906082062873,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0012,
|
|
"mean_token_accuracy": 0.11026087254285813,
|
|
"num_tokens": 231826.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.417343282699585,
|
|
"epoch": 0.00761122771409461,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9298,
|
|
"mean_token_accuracy": 0.11452125087380409,
|
|
"num_tokens": 235519.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.386319303512574,
|
|
"epoch": 0.007726549346126346,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1777,
|
|
"mean_token_accuracy": 0.10753077790141105,
|
|
"num_tokens": 239451.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.28514552116394,
|
|
"epoch": 0.007841870978158083,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0215,
|
|
"mean_token_accuracy": 0.1078951895236969,
|
|
"num_tokens": 243193.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.671026420593262,
|
|
"epoch": 0.007957192610189819,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0501,
|
|
"mean_token_accuracy": 0.11457417979836464,
|
|
"num_tokens": 246859.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.1775932788848875,
|
|
"epoch": 0.008072514242221556,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9368,
|
|
"mean_token_accuracy": 0.11339911893010139,
|
|
"num_tokens": 250345.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.443285751342773,
|
|
"epoch": 0.008187835874253292,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9842,
|
|
"mean_token_accuracy": 0.10223566293716431,
|
|
"num_tokens": 254201.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.363960456848145,
|
|
"epoch": 0.00830315750628503,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9875,
|
|
"mean_token_accuracy": 0.11335918605327606,
|
|
"num_tokens": 258073.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.3993799686431885,
|
|
"epoch": 0.008418479138316766,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9985,
|
|
"mean_token_accuracy": 0.1107051931321621,
|
|
"num_tokens": 262221.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.2681444644927975,
|
|
"epoch": 0.008533800770348501,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8422,
|
|
"mean_token_accuracy": 0.11727887839078903,
|
|
"num_tokens": 265920.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.339032983779907,
|
|
"epoch": 0.008649122402380239,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8632,
|
|
"mean_token_accuracy": 0.1098533809185028,
|
|
"num_tokens": 269617.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.235262298583985,
|
|
"epoch": 0.008764444034411975,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0343,
|
|
"mean_token_accuracy": 0.11555488631129265,
|
|
"num_tokens": 273482.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.33788275718689,
|
|
"epoch": 0.008879765666443712,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8574,
|
|
"mean_token_accuracy": 0.12360220775008202,
|
|
"num_tokens": 276877.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.284343814849853,
|
|
"epoch": 0.008995087298475448,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.1250680521130562,
|
|
"num_tokens": 280342.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.164630126953125,
|
|
"epoch": 0.009110408930507184,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8672,
|
|
"mean_token_accuracy": 0.11904213950037956,
|
|
"num_tokens": 283740.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.220073223114014,
|
|
"epoch": 0.009225730562538921,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8448,
|
|
"mean_token_accuracy": 0.114653430134058,
|
|
"num_tokens": 287534.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.279120349884034,
|
|
"epoch": 0.009341052194570657,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.9441,
|
|
"mean_token_accuracy": 0.11235265359282494,
|
|
"num_tokens": 290847.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.284884881973267,
|
|
"epoch": 0.009456373826602395,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8057,
|
|
"mean_token_accuracy": 0.11482253968715668,
|
|
"num_tokens": 294317.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.207005405426026,
|
|
"epoch": 0.00957169545863413,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7533,
|
|
"mean_token_accuracy": 0.11797683015465736,
|
|
"num_tokens": 297899.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.00657844543457,
|
|
"epoch": 0.009687017090665866,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.789,
|
|
"mean_token_accuracy": 0.1228546530008316,
|
|
"num_tokens": 301076.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.3213354587554935,
|
|
"epoch": 0.009802338722697604,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7636,
|
|
"mean_token_accuracy": 0.12574946507811546,
|
|
"num_tokens": 304848.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.14543981552124,
|
|
"epoch": 0.00991766035472934,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.8416,
|
|
"mean_token_accuracy": 0.11870098114013672,
|
|
"num_tokens": 308223.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.262237453460694,
|
|
"epoch": 0.010032981986761077,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.9775,
|
|
"mean_token_accuracy": 0.11694321408867836,
|
|
"num_tokens": 311755.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.160612010955811,
|
|
"epoch": 0.010148303618792813,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8402,
|
|
"mean_token_accuracy": 0.12193377017974853,
|
|
"num_tokens": 315362.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.3824732303619385,
|
|
"epoch": 0.010263625250824549,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0482,
|
|
"mean_token_accuracy": 0.10882357284426689,
|
|
"num_tokens": 319705.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.185869121551514,
|
|
"epoch": 0.010378946882856286,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8825,
|
|
"mean_token_accuracy": 0.11569961458444596,
|
|
"num_tokens": 323371.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.384743690490723,
|
|
"epoch": 0.010494268514888022,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.981,
|
|
"mean_token_accuracy": 0.11269342377781869,
|
|
"num_tokens": 326846.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.022752046585083,
|
|
"epoch": 0.01060959014691976,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7201,
|
|
"mean_token_accuracy": 0.13362340852618218,
|
|
"num_tokens": 330305.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.039274597167969,
|
|
"epoch": 0.010724911778951495,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7251,
|
|
"mean_token_accuracy": 0.1254519484937191,
|
|
"num_tokens": 333532.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.0497644424438475,
|
|
"epoch": 0.010840233410983233,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7665,
|
|
"mean_token_accuracy": 0.12782322838902474,
|
|
"num_tokens": 336892.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.166635799407959,
|
|
"epoch": 0.010955555043014969,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7096,
|
|
"mean_token_accuracy": 0.1333428256213665,
|
|
"num_tokens": 340372.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.2982118129730225,
|
|
"epoch": 0.011070876675046705,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9193,
|
|
"mean_token_accuracy": 0.1203345313668251,
|
|
"num_tokens": 344204.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.994412517547607,
|
|
"epoch": 0.011186198307078442,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7874,
|
|
"mean_token_accuracy": 0.1303581975400448,
|
|
"num_tokens": 347962.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.16865119934082,
|
|
"epoch": 0.011301519939110178,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7901,
|
|
"mean_token_accuracy": 0.12705896645784379,
|
|
"num_tokens": 351559.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.113827466964722,
|
|
"epoch": 0.011416841571141916,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8509,
|
|
"mean_token_accuracy": 0.11782361045479775,
|
|
"num_tokens": 354878.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.1837897300720215,
|
|
"epoch": 0.011532163203173651,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8408,
|
|
"mean_token_accuracy": 0.11947782784700393,
|
|
"num_tokens": 358998.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.008907651901245,
|
|
"epoch": 0.011647484835205387,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8216,
|
|
"mean_token_accuracy": 0.12814051583409308,
|
|
"num_tokens": 362372.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.99374098777771,
|
|
"epoch": 0.011762806467237125,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7065,
|
|
"mean_token_accuracy": 0.12169086709618568,
|
|
"num_tokens": 366142.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.006280326843262,
|
|
"epoch": 0.01187812809926886,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7488,
|
|
"mean_token_accuracy": 0.1289787232875824,
|
|
"num_tokens": 369761.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.978412103652954,
|
|
"epoch": 0.011993449731300598,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6808,
|
|
"mean_token_accuracy": 0.1272241547703743,
|
|
"num_tokens": 373145.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.046985721588134,
|
|
"epoch": 0.012108771363332334,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6475,
|
|
"mean_token_accuracy": 0.12157402634620666,
|
|
"num_tokens": 376447.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.014685773849488,
|
|
"epoch": 0.01222409299536407,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8095,
|
|
"mean_token_accuracy": 0.11625444144010544,
|
|
"num_tokens": 379928.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.930046272277832,
|
|
"epoch": 0.012339414627395807,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.7477,
|
|
"mean_token_accuracy": 0.11298542842268944,
|
|
"num_tokens": 383443.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.917828035354614,
|
|
"epoch": 0.012454736259427543,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5982,
|
|
"mean_token_accuracy": 0.1328738048672676,
|
|
"num_tokens": 386691.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.044282913208008,
|
|
"epoch": 0.01257005789145928,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8707,
|
|
"mean_token_accuracy": 0.12004515156149864,
|
|
"num_tokens": 390600.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.009495115280151,
|
|
"epoch": 0.012685379523491016,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.712,
|
|
"mean_token_accuracy": 0.11927132830023765,
|
|
"num_tokens": 394148.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.963443565368652,
|
|
"epoch": 0.012800701155522754,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6035,
|
|
"mean_token_accuracy": 0.13802491277456283,
|
|
"num_tokens": 397524.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.918949985504151,
|
|
"epoch": 0.01291602278755449,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7757,
|
|
"mean_token_accuracy": 0.1237744927406311,
|
|
"num_tokens": 401366.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.8452520847320555,
|
|
"epoch": 0.013031344419586225,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7075,
|
|
"mean_token_accuracy": 0.12387786731123925,
|
|
"num_tokens": 404952.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.006792783737183,
|
|
"epoch": 0.013146666051617963,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6857,
|
|
"mean_token_accuracy": 0.12056663036346435,
|
|
"num_tokens": 408678.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.869858264923096,
|
|
"epoch": 0.013261987683649699,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.737,
|
|
"mean_token_accuracy": 0.1358136773109436,
|
|
"num_tokens": 411964.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.9903857707977295,
|
|
"epoch": 0.013377309315681436,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7629,
|
|
"mean_token_accuracy": 0.12550165876746178,
|
|
"num_tokens": 415533.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.999003601074219,
|
|
"epoch": 0.013492630947713172,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7381,
|
|
"mean_token_accuracy": 0.11583735272288323,
|
|
"num_tokens": 419052.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.950546169281006,
|
|
"epoch": 0.013607952579744908,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8094,
|
|
"mean_token_accuracy": 0.11839277520775796,
|
|
"num_tokens": 422775.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.002314567565918,
|
|
"epoch": 0.013723274211776645,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7166,
|
|
"mean_token_accuracy": 0.12862658202648164,
|
|
"num_tokens": 426363.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.846304988861084,
|
|
"epoch": 0.013838595843808381,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7329,
|
|
"mean_token_accuracy": 0.13125129267573357,
|
|
"num_tokens": 429815.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.880826377868653,
|
|
"epoch": 0.013953917475840119,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6259,
|
|
"mean_token_accuracy": 0.13348380103707314,
|
|
"num_tokens": 433128.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.932210779190063,
|
|
"epoch": 0.014069239107871855,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6694,
|
|
"mean_token_accuracy": 0.12162100598216057,
|
|
"num_tokens": 436806.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.953799104690551,
|
|
"epoch": 0.01418456073990359,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7108,
|
|
"mean_token_accuracy": 0.13104304745793344,
|
|
"num_tokens": 440601.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.906950569152832,
|
|
"epoch": 0.014299882371935328,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7254,
|
|
"mean_token_accuracy": 0.12025304809212685,
|
|
"num_tokens": 444192.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.865710210800171,
|
|
"epoch": 0.014415204003967064,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.77,
|
|
"mean_token_accuracy": 0.1289124071598053,
|
|
"num_tokens": 447680.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.900669288635254,
|
|
"epoch": 0.014530525635998801,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7254,
|
|
"mean_token_accuracy": 0.12785449996590614,
|
|
"num_tokens": 451469.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.79181227684021,
|
|
"epoch": 0.014645847268030537,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5524,
|
|
"mean_token_accuracy": 0.13745752573013306,
|
|
"num_tokens": 455099.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.9072105407714846,
|
|
"epoch": 0.014761168900062273,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.6685,
|
|
"mean_token_accuracy": 0.1308851085603237,
|
|
"num_tokens": 459026.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.6932032108306885,
|
|
"epoch": 0.01487649053209401,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.607,
|
|
"mean_token_accuracy": 0.12305137664079666,
|
|
"num_tokens": 462901.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.998805093765259,
|
|
"epoch": 0.014991812164125746,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.76,
|
|
"mean_token_accuracy": 0.1286436066031456,
|
|
"num_tokens": 466424.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.72501015663147,
|
|
"epoch": 0.015107133796157484,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5726,
|
|
"mean_token_accuracy": 0.13623193502426148,
|
|
"num_tokens": 470063.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.730249643325806,
|
|
"epoch": 0.01522245542818922,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4954,
|
|
"mean_token_accuracy": 0.145492073148489,
|
|
"num_tokens": 473513.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.748109960556031,
|
|
"epoch": 0.015337777060220957,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5471,
|
|
"mean_token_accuracy": 0.1347307026386261,
|
|
"num_tokens": 476973.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.7143310546875,
|
|
"epoch": 0.015453098692252693,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6092,
|
|
"mean_token_accuracy": 0.12850266173481942,
|
|
"num_tokens": 480331.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.830945062637329,
|
|
"epoch": 0.015568420324284429,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5694,
|
|
"mean_token_accuracy": 0.11640301495790481,
|
|
"num_tokens": 483647.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.743711805343628,
|
|
"epoch": 0.015683741956316166,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.643,
|
|
"mean_token_accuracy": 0.1260574921965599,
|
|
"num_tokens": 487463.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.711105155944824,
|
|
"epoch": 0.015799063588347904,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5762,
|
|
"mean_token_accuracy": 0.13666252195835113,
|
|
"num_tokens": 490891.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.916413402557373,
|
|
"epoch": 0.015914385220379638,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.8476,
|
|
"mean_token_accuracy": 0.1245967648923397,
|
|
"num_tokens": 494576.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.638924217224121,
|
|
"epoch": 0.016029706852411375,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.571,
|
|
"mean_token_accuracy": 0.1299450919032097,
|
|
"num_tokens": 498316.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.887248802185058,
|
|
"epoch": 0.016145028484443113,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.642,
|
|
"mean_token_accuracy": 0.13247032091021538,
|
|
"num_tokens": 501746.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.567225503921509,
|
|
"epoch": 0.016260350116474847,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5599,
|
|
"mean_token_accuracy": 0.14109184890985488,
|
|
"num_tokens": 505188.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.827360010147094,
|
|
"epoch": 0.016375671748506584,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5304,
|
|
"mean_token_accuracy": 0.1347896784543991,
|
|
"num_tokens": 509342.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.785188245773315,
|
|
"epoch": 0.016490993380538322,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6174,
|
|
"mean_token_accuracy": 0.13861689269542693,
|
|
"num_tokens": 513284.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.648976516723633,
|
|
"epoch": 0.01660631501257006,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.637,
|
|
"mean_token_accuracy": 0.12227844670414925,
|
|
"num_tokens": 517032.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.8655952453613285,
|
|
"epoch": 0.016721636644601794,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5419,
|
|
"mean_token_accuracy": 0.14613815620541573,
|
|
"num_tokens": 520629.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.44150743484497,
|
|
"epoch": 0.01683695827663353,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4529,
|
|
"mean_token_accuracy": 0.14707944691181182,
|
|
"num_tokens": 524281.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.630420827865601,
|
|
"epoch": 0.01695227990866527,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5273,
|
|
"mean_token_accuracy": 0.1444172039628029,
|
|
"num_tokens": 527726.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.6363856315612795,
|
|
"epoch": 0.017067601540697003,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5555,
|
|
"mean_token_accuracy": 0.13690576776862146,
|
|
"num_tokens": 531017.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.761928415298462,
|
|
"epoch": 0.01718292317272874,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6378,
|
|
"mean_token_accuracy": 0.132551933825016,
|
|
"num_tokens": 535161.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.754014015197754,
|
|
"epoch": 0.017298244804760478,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5925,
|
|
"mean_token_accuracy": 0.1325087755918503,
|
|
"num_tokens": 538916.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.547227001190185,
|
|
"epoch": 0.017413566436792212,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.518,
|
|
"mean_token_accuracy": 0.13418273031711578,
|
|
"num_tokens": 542857.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.776309585571289,
|
|
"epoch": 0.01752888806882395,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6192,
|
|
"mean_token_accuracy": 0.12573943585157393,
|
|
"num_tokens": 546678.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.778603172302246,
|
|
"epoch": 0.017644209700855687,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5592,
|
|
"mean_token_accuracy": 0.12105903252959252,
|
|
"num_tokens": 550185.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.589523363113403,
|
|
"epoch": 0.017759531332887424,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5775,
|
|
"mean_token_accuracy": 0.12825108394026757,
|
|
"num_tokens": 553769.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.7237303256988525,
|
|
"epoch": 0.01787485296491916,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4662,
|
|
"mean_token_accuracy": 0.151670241355896,
|
|
"num_tokens": 557235.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.528875732421875,
|
|
"epoch": 0.017990174596950896,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.521,
|
|
"mean_token_accuracy": 0.14290796518325805,
|
|
"num_tokens": 560646.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.589083051681518,
|
|
"epoch": 0.018105496228982634,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.3976,
|
|
"mean_token_accuracy": 0.1500653974711895,
|
|
"num_tokens": 563945.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.734743690490722,
|
|
"epoch": 0.018220817861014368,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.6522,
|
|
"mean_token_accuracy": 0.13518640398979187,
|
|
"num_tokens": 568010.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.603312349319458,
|
|
"epoch": 0.018336139493046105,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5532,
|
|
"mean_token_accuracy": 0.13213913813233374,
|
|
"num_tokens": 571643.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.773268175125122,
|
|
"epoch": 0.018451461125077843,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4958,
|
|
"mean_token_accuracy": 0.13954881727695465,
|
|
"num_tokens": 575128.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.423713874816895,
|
|
"epoch": 0.01856678275710958,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4667,
|
|
"mean_token_accuracy": 0.13951935172080993,
|
|
"num_tokens": 578471.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.739634275436401,
|
|
"epoch": 0.018682104389141314,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4825,
|
|
"mean_token_accuracy": 0.1304561197757721,
|
|
"num_tokens": 581865.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.507706260681152,
|
|
"epoch": 0.018797426021173052,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5074,
|
|
"mean_token_accuracy": 0.13089399486780168,
|
|
"num_tokens": 585277.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.839502048492432,
|
|
"epoch": 0.01891274765320479,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.6384,
|
|
"mean_token_accuracy": 0.12499238029122353,
|
|
"num_tokens": 589313.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.49386773109436,
|
|
"epoch": 0.019028069285236524,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4789,
|
|
"mean_token_accuracy": 0.13701779171824455,
|
|
"num_tokens": 592907.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.685422277450561,
|
|
"epoch": 0.01914339091726826,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5011,
|
|
"mean_token_accuracy": 0.140823569893837,
|
|
"num_tokens": 596406.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.538355827331543,
|
|
"epoch": 0.0192587125493,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4467,
|
|
"mean_token_accuracy": 0.15094958543777465,
|
|
"num_tokens": 599978.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.474994850158692,
|
|
"epoch": 0.019374034181331733,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4284,
|
|
"mean_token_accuracy": 0.14493167251348496,
|
|
"num_tokens": 603466.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.6400146484375,
|
|
"epoch": 0.01948935581336347,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4499,
|
|
"mean_token_accuracy": 0.1442779131233692,
|
|
"num_tokens": 606976.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.627993726730347,
|
|
"epoch": 0.019604677445395208,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5223,
|
|
"mean_token_accuracy": 0.13213641494512557,
|
|
"num_tokens": 610475.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.567187404632568,
|
|
"epoch": 0.019719999077426945,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5697,
|
|
"mean_token_accuracy": 0.1341607928276062,
|
|
"num_tokens": 614416.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.646091604232788,
|
|
"epoch": 0.01983532070945868,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4594,
|
|
"mean_token_accuracy": 0.1447944924235344,
|
|
"num_tokens": 618398.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.448280954360962,
|
|
"epoch": 0.019950642341490417,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4029,
|
|
"mean_token_accuracy": 0.14234068989753723,
|
|
"num_tokens": 622029.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.480152940750122,
|
|
"epoch": 0.020065963973522154,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4506,
|
|
"mean_token_accuracy": 0.14138664305210114,
|
|
"num_tokens": 625438.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.640823078155518,
|
|
"epoch": 0.02018128560555389,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3926,
|
|
"mean_token_accuracy": 0.14264481738209725,
|
|
"num_tokens": 628936.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.582769060134888,
|
|
"epoch": 0.020296607237585626,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4521,
|
|
"mean_token_accuracy": 0.15119002610445023,
|
|
"num_tokens": 632190.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.552495670318604,
|
|
"epoch": 0.020411928869617364,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.5234,
|
|
"mean_token_accuracy": 0.1364043727517128,
|
|
"num_tokens": 635786.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.474388265609742,
|
|
"epoch": 0.020527250501649098,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4413,
|
|
"mean_token_accuracy": 0.14595962017774583,
|
|
"num_tokens": 639423.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.593573379516601,
|
|
"epoch": 0.020642572133680835,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4048,
|
|
"mean_token_accuracy": 0.14657237753272057,
|
|
"num_tokens": 642880.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.289942502975464,
|
|
"epoch": 0.020757893765712573,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4328,
|
|
"mean_token_accuracy": 0.1434159643948078,
|
|
"num_tokens": 646417.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.628101491928101,
|
|
"epoch": 0.02087321539774431,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4526,
|
|
"mean_token_accuracy": 0.13669708520174026,
|
|
"num_tokens": 650088.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.6030834197998045,
|
|
"epoch": 0.020988537029776044,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5712,
|
|
"mean_token_accuracy": 0.13861860781908036,
|
|
"num_tokens": 653648.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.652601528167724,
|
|
"epoch": 0.021103858661807782,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.6468,
|
|
"mean_token_accuracy": 0.1381187282502651,
|
|
"num_tokens": 657597.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.542675542831421,
|
|
"epoch": 0.02121918029383952,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5008,
|
|
"mean_token_accuracy": 0.13693511635065078,
|
|
"num_tokens": 661466.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.588531160354615,
|
|
"epoch": 0.021334501925871253,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5588,
|
|
"mean_token_accuracy": 0.1473455861210823,
|
|
"num_tokens": 665662.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.489068174362183,
|
|
"epoch": 0.02144982355790299,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3411,
|
|
"mean_token_accuracy": 0.1503666989505291,
|
|
"num_tokens": 669190.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.479677772521972,
|
|
"epoch": 0.02156514518993473,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4155,
|
|
"mean_token_accuracy": 0.13933071866631508,
|
|
"num_tokens": 672785.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.328510761260986,
|
|
"epoch": 0.021680466821966466,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3937,
|
|
"mean_token_accuracy": 0.13994476646184922,
|
|
"num_tokens": 676090.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.570944738388062,
|
|
"epoch": 0.0217957884539982,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4825,
|
|
"mean_token_accuracy": 0.13967012241482735,
|
|
"num_tokens": 679765.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.453943824768066,
|
|
"epoch": 0.021911110086029938,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4992,
|
|
"mean_token_accuracy": 0.13961242511868477,
|
|
"num_tokens": 683289.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.530073261260986,
|
|
"epoch": 0.022026431718061675,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3172,
|
|
"mean_token_accuracy": 0.14913771972060202,
|
|
"num_tokens": 686551.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.444345617294312,
|
|
"epoch": 0.02214175335009341,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5807,
|
|
"mean_token_accuracy": 0.13834290951490402,
|
|
"num_tokens": 689876.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.5710207462310795,
|
|
"epoch": 0.022257074982125147,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4255,
|
|
"mean_token_accuracy": 0.14252933487296104,
|
|
"num_tokens": 693258.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.419457006454468,
|
|
"epoch": 0.022372396614156884,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4773,
|
|
"mean_token_accuracy": 0.14677356630563737,
|
|
"num_tokens": 696908.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.683896541595459,
|
|
"epoch": 0.02248771824618862,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.5156,
|
|
"mean_token_accuracy": 0.1473010540008545,
|
|
"num_tokens": 700621.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.293711614608765,
|
|
"epoch": 0.022603039878220356,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3864,
|
|
"mean_token_accuracy": 0.14675331562757493,
|
|
"num_tokens": 703896.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.480734252929688,
|
|
"epoch": 0.022718361510252093,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2061,
|
|
"mean_token_accuracy": 0.15426455512642862,
|
|
"num_tokens": 707309.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.292126560211182,
|
|
"epoch": 0.02283368314228383,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3617,
|
|
"mean_token_accuracy": 0.14406633898615837,
|
|
"num_tokens": 710719.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.513804578781128,
|
|
"epoch": 0.022949004774315565,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.534,
|
|
"mean_token_accuracy": 0.14377237409353255,
|
|
"num_tokens": 715016.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.560623693466186,
|
|
"epoch": 0.023064326406347303,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.51,
|
|
"mean_token_accuracy": 0.13352807313203813,
|
|
"num_tokens": 718863.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.375119304656982,
|
|
"epoch": 0.02317964803837904,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.3054,
|
|
"mean_token_accuracy": 0.1605803407728672,
|
|
"num_tokens": 722399.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.316158056259155,
|
|
"epoch": 0.023294969670410774,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.405,
|
|
"mean_token_accuracy": 0.1526450350880623,
|
|
"num_tokens": 726090.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.456164693832397,
|
|
"epoch": 0.023410291302442512,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.3535,
|
|
"mean_token_accuracy": 0.14333662390708923,
|
|
"num_tokens": 729749.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.468470811843872,
|
|
"epoch": 0.02352561293447425,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.4095,
|
|
"mean_token_accuracy": 0.14930933490395545,
|
|
"num_tokens": 733370.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.43420557975769,
|
|
"epoch": 0.023640934566505987,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.4476,
|
|
"mean_token_accuracy": 0.1455454170703888,
|
|
"num_tokens": 737084.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.435062837600708,
|
|
"epoch": 0.02375625619853772,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.465,
|
|
"mean_token_accuracy": 0.13956179320812226,
|
|
"num_tokens": 740907.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.49959626197815,
|
|
"epoch": 0.02387157783056946,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.3637,
|
|
"mean_token_accuracy": 0.139167670160532,
|
|
"num_tokens": 744546.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.455142593383789,
|
|
"epoch": 0.023986899462601196,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.5537,
|
|
"mean_token_accuracy": 0.14254215210676194,
|
|
"num_tokens": 748289.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.4490211486816404,
|
|
"epoch": 0.02410222109463293,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.3131,
|
|
"mean_token_accuracy": 0.1561379536986351,
|
|
"num_tokens": 752034.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.270240736007691,
|
|
"epoch": 0.024217542726664668,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.2185,
|
|
"mean_token_accuracy": 0.14977468624711038,
|
|
"num_tokens": 755312.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.373951244354248,
|
|
"epoch": 0.024332864358696405,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.3739,
|
|
"mean_token_accuracy": 0.14269247725605966,
|
|
"num_tokens": 758921.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.298541164398193,
|
|
"epoch": 0.02444818599072814,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.3118,
|
|
"mean_token_accuracy": 0.14421494603157042,
|
|
"num_tokens": 762392.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.290153551101684,
|
|
"epoch": 0.024563507622759877,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.2757,
|
|
"mean_token_accuracy": 0.1500731386244297,
|
|
"num_tokens": 766151.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.411606025695801,
|
|
"epoch": 0.024678829254791614,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.4063,
|
|
"mean_token_accuracy": 0.13883894458413124,
|
|
"num_tokens": 769799.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.339744710922242,
|
|
"epoch": 0.024794150886823352,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.3998,
|
|
"mean_token_accuracy": 0.1386982448399067,
|
|
"num_tokens": 773345.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.344364261627197,
|
|
"epoch": 0.024909472518855086,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.2794,
|
|
"mean_token_accuracy": 0.14550025388598442,
|
|
"num_tokens": 777121.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.332255935668945,
|
|
"epoch": 0.025024794150886823,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.2817,
|
|
"mean_token_accuracy": 0.15340236723423004,
|
|
"num_tokens": 780601.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.268877267837524,
|
|
"epoch": 0.02514011578291856,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.4835,
|
|
"mean_token_accuracy": 0.14280915558338164,
|
|
"num_tokens": 783995.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.43036732673645,
|
|
"epoch": 0.025255437414950295,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.3891,
|
|
"mean_token_accuracy": 0.14359135255217553,
|
|
"num_tokens": 787810.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.499591445922851,
|
|
"epoch": 0.025370759046982033,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.464,
|
|
"mean_token_accuracy": 0.1363844908773899,
|
|
"num_tokens": 791554.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.479581546783447,
|
|
"epoch": 0.02548608067901377,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.2954,
|
|
"mean_token_accuracy": 0.14641399681568146,
|
|
"num_tokens": 795210.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.109205007553101,
|
|
"epoch": 0.025601402311045508,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.2189,
|
|
"mean_token_accuracy": 0.15632987767457962,
|
|
"num_tokens": 798225.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.358191633224488,
|
|
"epoch": 0.02571672394307724,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.2967,
|
|
"mean_token_accuracy": 0.16011278629302977,
|
|
"num_tokens": 801465.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.34049768447876,
|
|
"epoch": 0.02583204557510898,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.2762,
|
|
"mean_token_accuracy": 0.15843751579523085,
|
|
"num_tokens": 804840.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.090512323379516,
|
|
"epoch": 0.025947367207140717,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.2974,
|
|
"mean_token_accuracy": 0.15075689256191255,
|
|
"num_tokens": 808594.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.4936192512512205,
|
|
"epoch": 0.02606268883917245,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.3657,
|
|
"mean_token_accuracy": 0.1478181280195713,
|
|
"num_tokens": 812495.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.5022705078125,
|
|
"epoch": 0.02617801047120419,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.4343,
|
|
"mean_token_accuracy": 0.1474046677350998,
|
|
"num_tokens": 816424.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.349539804458618,
|
|
"epoch": 0.026293332103235926,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.4422,
|
|
"mean_token_accuracy": 0.1453184999525547,
|
|
"num_tokens": 820757.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.458282709121704,
|
|
"epoch": 0.02640865373526766,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.3994,
|
|
"mean_token_accuracy": 0.1382867231965065,
|
|
"num_tokens": 824456.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.350621223449707,
|
|
"epoch": 0.026523975367299397,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.3269,
|
|
"mean_token_accuracy": 0.1577729970216751,
|
|
"num_tokens": 828144.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.164858722686768,
|
|
"epoch": 0.026639296999331135,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.2701,
|
|
"mean_token_accuracy": 0.15525535494089127,
|
|
"num_tokens": 831802.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.359017467498779,
|
|
"epoch": 0.026754618631362873,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.3614,
|
|
"mean_token_accuracy": 0.15466616973280906,
|
|
"num_tokens": 835355.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.398327922821045,
|
|
"epoch": 0.026869940263394607,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.4092,
|
|
"mean_token_accuracy": 0.14570969864726066,
|
|
"num_tokens": 839193.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.252834224700928,
|
|
"epoch": 0.026985261895426344,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.2631,
|
|
"mean_token_accuracy": 0.14109885692596436,
|
|
"num_tokens": 842601.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.2718846797943115,
|
|
"epoch": 0.02710058352745808,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.2755,
|
|
"mean_token_accuracy": 0.15494223684072495,
|
|
"num_tokens": 846335.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.341091966629028,
|
|
"epoch": 0.027215905159489816,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.2403,
|
|
"mean_token_accuracy": 0.1602032884955406,
|
|
"num_tokens": 849944.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.244535970687866,
|
|
"epoch": 0.027331226791521553,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.2533,
|
|
"mean_token_accuracy": 0.15198892951011658,
|
|
"num_tokens": 853417.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.322394323348999,
|
|
"epoch": 0.02744654842355329,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.3589,
|
|
"mean_token_accuracy": 0.1436021901667118,
|
|
"num_tokens": 856935.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.405830144882202,
|
|
"epoch": 0.02756187005558503,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.14548677504062651,
|
|
"num_tokens": 860334.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.394810914993286,
|
|
"epoch": 0.027677191687616762,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.2798,
|
|
"mean_token_accuracy": 0.15825977325439453,
|
|
"num_tokens": 864041.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.256173610687256,
|
|
"epoch": 0.0277925133196485,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.3102,
|
|
"mean_token_accuracy": 0.1614562138915062,
|
|
"num_tokens": 867577.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.306905031204224,
|
|
"epoch": 0.027907834951680238,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.329,
|
|
"mean_token_accuracy": 0.15291077867150307,
|
|
"num_tokens": 871509.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.388304615020752,
|
|
"epoch": 0.02802315658371197,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.2867,
|
|
"mean_token_accuracy": 0.16107984781265258,
|
|
"num_tokens": 874851.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.368912839889527,
|
|
"epoch": 0.02813847821574371,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.3097,
|
|
"mean_token_accuracy": 0.16125393360853196,
|
|
"num_tokens": 878516.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.301360034942627,
|
|
"epoch": 0.028253799847775447,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.2247,
|
|
"mean_token_accuracy": 0.1498270332813263,
|
|
"num_tokens": 882339.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.385891532897949,
|
|
"epoch": 0.02836912147980718,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.4883,
|
|
"mean_token_accuracy": 0.15265526697039605,
|
|
"num_tokens": 886439.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.418463468551636,
|
|
"epoch": 0.028484443111838918,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.3312,
|
|
"mean_token_accuracy": 0.1584172047674656,
|
|
"num_tokens": 890247.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.332312297821045,
|
|
"epoch": 0.028599764743870656,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.3083,
|
|
"mean_token_accuracy": 0.14334525242447854,
|
|
"num_tokens": 893576.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.410537958145142,
|
|
"epoch": 0.028715086375902393,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.3741,
|
|
"mean_token_accuracy": 0.14212653860449792,
|
|
"num_tokens": 897163.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.35398645401001,
|
|
"epoch": 0.028830408007934127,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.2691,
|
|
"mean_token_accuracy": 0.15415772050619125,
|
|
"num_tokens": 900787.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.258543062210083,
|
|
"epoch": 0.028945729639965865,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.2836,
|
|
"mean_token_accuracy": 0.15877098292112352,
|
|
"num_tokens": 904420.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.415359115600586,
|
|
"epoch": 0.029061051271997602,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.4341,
|
|
"mean_token_accuracy": 0.1414051905274391,
|
|
"num_tokens": 908244.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.351872682571411,
|
|
"epoch": 0.029176372904029337,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.238,
|
|
"mean_token_accuracy": 0.15497701838612557,
|
|
"num_tokens": 911715.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.099112224578858,
|
|
"epoch": 0.029291694536061074,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.1431,
|
|
"mean_token_accuracy": 0.17191263288259506,
|
|
"num_tokens": 915039.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.316360187530518,
|
|
"epoch": 0.02940701616809281,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.1564,
|
|
"mean_token_accuracy": 0.16846123561263085,
|
|
"num_tokens": 918744.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.324119234085083,
|
|
"epoch": 0.029522337800124546,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.321,
|
|
"mean_token_accuracy": 0.15262760668992997,
|
|
"num_tokens": 922235.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.636838054656982,
|
|
"epoch": 0.029637659432156283,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.3354,
|
|
"mean_token_accuracy": 0.1518080860376358,
|
|
"num_tokens": 926282.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.4565459251403805,
|
|
"epoch": 0.02975298106418802,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.4512,
|
|
"mean_token_accuracy": 0.14075498580932616,
|
|
"num_tokens": 930226.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.220556306838989,
|
|
"epoch": 0.02986830269621976,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.1541,
|
|
"mean_token_accuracy": 0.16578194499015808,
|
|
"num_tokens": 933334.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.208025884628296,
|
|
"epoch": 0.029983624328251492,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.1323,
|
|
"mean_token_accuracy": 0.1657858334481716,
|
|
"num_tokens": 936710.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.213149499893189,
|
|
"epoch": 0.03009894596028323,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.2428,
|
|
"mean_token_accuracy": 0.15492171049118042,
|
|
"num_tokens": 940045.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.254480504989624,
|
|
"epoch": 0.030214267592314967,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.1529,
|
|
"mean_token_accuracy": 0.1686461329460144,
|
|
"num_tokens": 943244.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.283830785751343,
|
|
"epoch": 0.0303295892243467,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.2155,
|
|
"mean_token_accuracy": 0.1567380040884018,
|
|
"num_tokens": 946590.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.297065305709839,
|
|
"epoch": 0.03044491085637844,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.1026,
|
|
"mean_token_accuracy": 0.1661017879843712,
|
|
"num_tokens": 950027.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.0860552310943605,
|
|
"epoch": 0.030560232488410177,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.2072,
|
|
"mean_token_accuracy": 0.16063485741615297,
|
|
"num_tokens": 953542.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.231730651855469,
|
|
"epoch": 0.030675554120441914,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.1028,
|
|
"mean_token_accuracy": 0.16824122667312622,
|
|
"num_tokens": 957031.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.269879579544067,
|
|
"epoch": 0.030790875752473648,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.2613,
|
|
"mean_token_accuracy": 0.15151529535651206,
|
|
"num_tokens": 960895.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.162332487106323,
|
|
"epoch": 0.030906197384505386,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.153,
|
|
"mean_token_accuracy": 0.157187882065773,
|
|
"num_tokens": 964532.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.1036382675170895,
|
|
"epoch": 0.031021519016537123,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.1991,
|
|
"mean_token_accuracy": 0.15546992048621178,
|
|
"num_tokens": 968256.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.298920583724976,
|
|
"epoch": 0.031136840648568857,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.0824,
|
|
"mean_token_accuracy": 0.17230783998966218,
|
|
"num_tokens": 971626.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.091312885284424,
|
|
"epoch": 0.0312521622806006,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.1444,
|
|
"mean_token_accuracy": 0.16052730455994607,
|
|
"num_tokens": 975116.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.269405841827393,
|
|
"epoch": 0.03136748391263233,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.2707,
|
|
"mean_token_accuracy": 0.14912461563944818,
|
|
"num_tokens": 978899.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.383816862106324,
|
|
"epoch": 0.031482805544664066,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.3754,
|
|
"mean_token_accuracy": 0.14807838052511216,
|
|
"num_tokens": 982993.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.269737339019775,
|
|
"epoch": 0.03159812717669581,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.0675,
|
|
"mean_token_accuracy": 0.16973973140120507,
|
|
"num_tokens": 986580.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.025365924835205,
|
|
"epoch": 0.03171344880872754,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.0606,
|
|
"mean_token_accuracy": 0.1691105842590332,
|
|
"num_tokens": 990248.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.094328594207764,
|
|
"epoch": 0.031828770440759276,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.1345,
|
|
"mean_token_accuracy": 0.14461168348789216,
|
|
"num_tokens": 994054.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.166022777557373,
|
|
"epoch": 0.03194409207279102,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.241,
|
|
"mean_token_accuracy": 0.15347518026828766,
|
|
"num_tokens": 997618.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.339942169189453,
|
|
"epoch": 0.03205941370482275,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.2378,
|
|
"mean_token_accuracy": 0.14919298738241196,
|
|
"num_tokens": 1001715.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.163516664505005,
|
|
"epoch": 0.032174735336854485,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.1914,
|
|
"mean_token_accuracy": 0.16769849807024,
|
|
"num_tokens": 1005405.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.128510999679565,
|
|
"epoch": 0.032290056968886226,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.0731,
|
|
"mean_token_accuracy": 0.17466477155685425,
|
|
"num_tokens": 1008892.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.176671314239502,
|
|
"epoch": 0.03240537860091796,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.2838,
|
|
"mean_token_accuracy": 0.15127266347408294,
|
|
"num_tokens": 1012658.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.311561632156372,
|
|
"epoch": 0.032520700232949694,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.237,
|
|
"mean_token_accuracy": 0.16338927447795867,
|
|
"num_tokens": 1016187.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.333290863037109,
|
|
"epoch": 0.032636021864981435,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.1761,
|
|
"mean_token_accuracy": 0.1574217602610588,
|
|
"num_tokens": 1019598.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.14404034614563,
|
|
"epoch": 0.03275134349701317,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.1026,
|
|
"mean_token_accuracy": 0.16011720150709152,
|
|
"num_tokens": 1022948.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.070792579650879,
|
|
"epoch": 0.0328666651290449,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.1454,
|
|
"mean_token_accuracy": 0.1638325572013855,
|
|
"num_tokens": 1026376.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.176722717285156,
|
|
"epoch": 0.032981986761076644,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.2639,
|
|
"mean_token_accuracy": 0.16633117124438285,
|
|
"num_tokens": 1029882.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.31840033531189,
|
|
"epoch": 0.03309730839310838,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.2051,
|
|
"mean_token_accuracy": 0.15748205184936523,
|
|
"num_tokens": 1033663.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.185667896270752,
|
|
"epoch": 0.03321263002514012,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.2437,
|
|
"mean_token_accuracy": 0.15791511163115501,
|
|
"num_tokens": 1037250.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.245442581176758,
|
|
"epoch": 0.03332795165717185,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.3109,
|
|
"mean_token_accuracy": 0.14368694126605988,
|
|
"num_tokens": 1040848.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.171295642852783,
|
|
"epoch": 0.03344327328920359,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.2145,
|
|
"mean_token_accuracy": 0.16131526231765747,
|
|
"num_tokens": 1044268.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.370648002624511,
|
|
"epoch": 0.03355859492123533,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.2616,
|
|
"mean_token_accuracy": 0.14988441914319992,
|
|
"num_tokens": 1048374.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.085245418548584,
|
|
"epoch": 0.03367391655326706,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.0067,
|
|
"mean_token_accuracy": 0.16004470884799957,
|
|
"num_tokens": 1051799.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.220491123199463,
|
|
"epoch": 0.033789238185298796,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.1373,
|
|
"mean_token_accuracy": 0.155309996008873,
|
|
"num_tokens": 1055460.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.179386472702026,
|
|
"epoch": 0.03390455981733054,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.1541,
|
|
"mean_token_accuracy": 0.15267691016197205,
|
|
"num_tokens": 1058995.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.116367673873901,
|
|
"epoch": 0.03401988144936227,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.1515,
|
|
"mean_token_accuracy": 0.1600068524479866,
|
|
"num_tokens": 1062631.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.112538814544678,
|
|
"epoch": 0.034135203081394005,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.0725,
|
|
"mean_token_accuracy": 0.16084999293088914,
|
|
"num_tokens": 1065671.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.122194862365722,
|
|
"epoch": 0.034250524713425747,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.1088,
|
|
"mean_token_accuracy": 0.16508205682039262,
|
|
"num_tokens": 1069123.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.300327444076538,
|
|
"epoch": 0.03436584634545748,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.2065,
|
|
"mean_token_accuracy": 0.15478258579969406,
|
|
"num_tokens": 1073063.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.214800214767456,
|
|
"epoch": 0.034481167977489215,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.196,
|
|
"mean_token_accuracy": 0.1586817130446434,
|
|
"num_tokens": 1076914.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.207804536819458,
|
|
"epoch": 0.034596489609520956,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.2121,
|
|
"mean_token_accuracy": 0.14832273870706558,
|
|
"num_tokens": 1080422.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.299990606307984,
|
|
"epoch": 0.03471181124155269,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.2456,
|
|
"mean_token_accuracy": 0.15910418182611466,
|
|
"num_tokens": 1084000.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.215333127975464,
|
|
"epoch": 0.034827132873584424,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.1652,
|
|
"mean_token_accuracy": 0.1699425533413887,
|
|
"num_tokens": 1087296.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.305229234695434,
|
|
"epoch": 0.034942454505616165,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.0757,
|
|
"mean_token_accuracy": 0.16299125403165818,
|
|
"num_tokens": 1090868.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.071372175216675,
|
|
"epoch": 0.0350577761376479,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.1249,
|
|
"mean_token_accuracy": 0.15510240346193313,
|
|
"num_tokens": 1094940.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.298955678939819,
|
|
"epoch": 0.03517309776967964,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.1248,
|
|
"mean_token_accuracy": 0.149101385474205,
|
|
"num_tokens": 1098556.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.203650283813476,
|
|
"epoch": 0.035288419401711374,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.2417,
|
|
"mean_token_accuracy": 0.15527782142162322,
|
|
"num_tokens": 1102433.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.0423609733581545,
|
|
"epoch": 0.03540374103374311,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.1441,
|
|
"mean_token_accuracy": 0.16072777137160302,
|
|
"num_tokens": 1105919.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.308422613143921,
|
|
"epoch": 0.03551906266577485,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.2218,
|
|
"mean_token_accuracy": 0.16015318781137466,
|
|
"num_tokens": 1109756.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.231192970275879,
|
|
"epoch": 0.03563438429780658,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.0422,
|
|
"mean_token_accuracy": 0.16985733211040496,
|
|
"num_tokens": 1113199.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.132574415206909,
|
|
"epoch": 0.03574970592983832,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.1938,
|
|
"mean_token_accuracy": 0.15479369163513185,
|
|
"num_tokens": 1116834.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.309852647781372,
|
|
"epoch": 0.03586502756187006,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.1743,
|
|
"mean_token_accuracy": 0.15377985835075378,
|
|
"num_tokens": 1120429.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.207673454284668,
|
|
"epoch": 0.03598034919390179,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.1465,
|
|
"mean_token_accuracy": 0.16769928634166717,
|
|
"num_tokens": 1123775.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.178130340576172,
|
|
"epoch": 0.036095670825933526,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.1925,
|
|
"mean_token_accuracy": 0.15238797962665557,
|
|
"num_tokens": 1127752.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.094422721862793,
|
|
"epoch": 0.03621099245796527,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 5.9932,
|
|
"mean_token_accuracy": 0.17406838685274123,
|
|
"num_tokens": 1131238.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.118645095825196,
|
|
"epoch": 0.036326314089997,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.1233,
|
|
"mean_token_accuracy": 0.15426641553640366,
|
|
"num_tokens": 1134882.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.239434480667114,
|
|
"epoch": 0.036441635722028735,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.0646,
|
|
"mean_token_accuracy": 0.1605118915438652,
|
|
"num_tokens": 1138410.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.164656639099121,
|
|
"epoch": 0.036556957354060476,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.0511,
|
|
"mean_token_accuracy": 0.16745497435331344,
|
|
"num_tokens": 1141737.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.164981126785278,
|
|
"epoch": 0.03667227898609221,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.1403,
|
|
"mean_token_accuracy": 0.16387298405170442,
|
|
"num_tokens": 1145647.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.176317310333252,
|
|
"epoch": 0.036787600618123945,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.2386,
|
|
"mean_token_accuracy": 0.1529179185628891,
|
|
"num_tokens": 1149546.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.277850675582886,
|
|
"epoch": 0.036902922250155686,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 5.922,
|
|
"mean_token_accuracy": 0.17315406799316407,
|
|
"num_tokens": 1152816.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.904309701919556,
|
|
"epoch": 0.03701824388218742,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 5.9656,
|
|
"mean_token_accuracy": 0.1816538780927658,
|
|
"num_tokens": 1155993.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.077587556838989,
|
|
"epoch": 0.03713356551421916,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.0792,
|
|
"mean_token_accuracy": 0.16475852727890014,
|
|
"num_tokens": 1159497.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.0507410049438475,
|
|
"epoch": 0.037248887146250895,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.0242,
|
|
"mean_token_accuracy": 0.1662035495042801,
|
|
"num_tokens": 1163040.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.245137310028076,
|
|
"epoch": 0.03736420877828263,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.1839,
|
|
"mean_token_accuracy": 0.1598523199558258,
|
|
"num_tokens": 1166645.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.327967691421509,
|
|
"epoch": 0.03747953041031437,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.171,
|
|
"mean_token_accuracy": 0.16097911745309829,
|
|
"num_tokens": 1170178.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.104309940338135,
|
|
"epoch": 0.037594852042346104,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.1204,
|
|
"mean_token_accuracy": 0.15574230700731279,
|
|
"num_tokens": 1173773.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.1969810962677006,
|
|
"epoch": 0.03771017367437784,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.1673,
|
|
"mean_token_accuracy": 0.15242768377065657,
|
|
"num_tokens": 1177233.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.174960088729859,
|
|
"epoch": 0.03782549530640958,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 5.9957,
|
|
"mean_token_accuracy": 0.16600930839776992,
|
|
"num_tokens": 1180870.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.910323619842529,
|
|
"epoch": 0.03794081693844131,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 5.9591,
|
|
"mean_token_accuracy": 0.17246160358190538,
|
|
"num_tokens": 1184313.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.0432041645050045,
|
|
"epoch": 0.03805613857047305,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.0782,
|
|
"mean_token_accuracy": 0.1528872460126877,
|
|
"num_tokens": 1187743.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.070108795166016,
|
|
"epoch": 0.03817146020250479,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 5.99,
|
|
"mean_token_accuracy": 0.174486380815506,
|
|
"num_tokens": 1191486.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.043355989456177,
|
|
"epoch": 0.03828678183453652,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.0193,
|
|
"mean_token_accuracy": 0.16831168085336684,
|
|
"num_tokens": 1194668.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.094351291656494,
|
|
"epoch": 0.038402103466568256,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.1042,
|
|
"mean_token_accuracy": 0.1645447552204132,
|
|
"num_tokens": 1198465.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.099325037002563,
|
|
"epoch": 0.0385174250986,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.0878,
|
|
"mean_token_accuracy": 0.1684389069676399,
|
|
"num_tokens": 1202343.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.160552263259888,
|
|
"epoch": 0.03863274673063173,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.0271,
|
|
"mean_token_accuracy": 0.16442104130983354,
|
|
"num_tokens": 1205680.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.924587535858154,
|
|
"epoch": 0.038748068362663465,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.18062706142663956,
|
|
"num_tokens": 1209105.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.096160173416138,
|
|
"epoch": 0.038863389994695206,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.15812493711709977,
|
|
"num_tokens": 1212485.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.1795238018035885,
|
|
"epoch": 0.03897871162672694,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.0774,
|
|
"mean_token_accuracy": 0.1687271252274513,
|
|
"num_tokens": 1216052.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.114777231216431,
|
|
"epoch": 0.03909403325875868,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.1188,
|
|
"mean_token_accuracy": 0.16740818470716476,
|
|
"num_tokens": 1219542.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.028544569015503,
|
|
"epoch": 0.039209354890790415,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.0876,
|
|
"mean_token_accuracy": 0.1561695158481598,
|
|
"num_tokens": 1223243.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.237950801849365,
|
|
"epoch": 0.03932467652282215,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.0873,
|
|
"mean_token_accuracy": 0.165063014626503,
|
|
"num_tokens": 1226764.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.1528244972229,
|
|
"epoch": 0.03943999815485389,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.0627,
|
|
"mean_token_accuracy": 0.166194324195385,
|
|
"num_tokens": 1230361.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.019901323318481,
|
|
"epoch": 0.039555319786885625,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.1796,
|
|
"mean_token_accuracy": 0.1515389233827591,
|
|
"num_tokens": 1233789.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.107355642318725,
|
|
"epoch": 0.03967064141891736,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.0243,
|
|
"mean_token_accuracy": 0.16518253833055496,
|
|
"num_tokens": 1237552.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.175836133956909,
|
|
"epoch": 0.0397859630509491,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.0467,
|
|
"mean_token_accuracy": 0.1534863218665123,
|
|
"num_tokens": 1241273.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.097687816619873,
|
|
"epoch": 0.039901284682980834,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.1526,
|
|
"mean_token_accuracy": 0.1575276270508766,
|
|
"num_tokens": 1245072.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.091709899902344,
|
|
"epoch": 0.04001660631501257,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.012,
|
|
"mean_token_accuracy": 0.170663982629776,
|
|
"num_tokens": 1248744.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.073425054550171,
|
|
"epoch": 0.04013192794704431,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.0474,
|
|
"mean_token_accuracy": 0.16608797684311866,
|
|
"num_tokens": 1252373.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.11737289428711,
|
|
"epoch": 0.04024724957907604,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.9772,
|
|
"mean_token_accuracy": 0.1798662781715393,
|
|
"num_tokens": 1255866.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.090176010131836,
|
|
"epoch": 0.04036257121110778,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.0869,
|
|
"mean_token_accuracy": 0.1616697147488594,
|
|
"num_tokens": 1259456.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.019082117080688,
|
|
"epoch": 0.04047789284313952,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.0108,
|
|
"mean_token_accuracy": 0.16962203308939933,
|
|
"num_tokens": 1263038.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.118891191482544,
|
|
"epoch": 0.04059321447517125,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.0384,
|
|
"mean_token_accuracy": 0.174375718832016,
|
|
"num_tokens": 1266905.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.169350337982178,
|
|
"epoch": 0.040708536107202986,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.0951,
|
|
"mean_token_accuracy": 0.16224005371332167,
|
|
"num_tokens": 1270614.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.0967010974884035,
|
|
"epoch": 0.04082385773923473,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 5.9841,
|
|
"mean_token_accuracy": 0.1610086090862751,
|
|
"num_tokens": 1274214.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.121833086013794,
|
|
"epoch": 0.04093917937126646,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.18188032805919646,
|
|
"num_tokens": 1277611.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.9284069538116455,
|
|
"epoch": 0.041054501003298195,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 5.9686,
|
|
"mean_token_accuracy": 0.16471225768327713,
|
|
"num_tokens": 1281168.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.11559944152832,
|
|
"epoch": 0.041169822635329936,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.0756,
|
|
"mean_token_accuracy": 0.16965191811323166,
|
|
"num_tokens": 1284869.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.198210525512695,
|
|
"epoch": 0.04128514426736167,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.0814,
|
|
"mean_token_accuracy": 0.1581938832998276,
|
|
"num_tokens": 1288666.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.01418023109436,
|
|
"epoch": 0.04140046589939341,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.1123,
|
|
"mean_token_accuracy": 0.1629746027290821,
|
|
"num_tokens": 1292139.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.248403644561767,
|
|
"epoch": 0.041515787531425145,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.036,
|
|
"mean_token_accuracy": 0.15748263150453568,
|
|
"num_tokens": 1295951.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.098617649078369,
|
|
"epoch": 0.04163110916345688,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.0489,
|
|
"mean_token_accuracy": 0.1594049036502838,
|
|
"num_tokens": 1299512.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.977329349517822,
|
|
"epoch": 0.04174643079548862,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 5.999,
|
|
"mean_token_accuracy": 0.16566509902477264,
|
|
"num_tokens": 1303159.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.179058933258057,
|
|
"epoch": 0.041861752427520355,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.1168,
|
|
"mean_token_accuracy": 0.16273373067379,
|
|
"num_tokens": 1307109.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.222625970840454,
|
|
"epoch": 0.04197707405955209,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.0088,
|
|
"mean_token_accuracy": 0.1648745894432068,
|
|
"num_tokens": 1310742.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.1200212955474855,
|
|
"epoch": 0.04209239569158383,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.2173,
|
|
"mean_token_accuracy": 0.16236369609832763,
|
|
"num_tokens": 1314321.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.064646816253662,
|
|
"epoch": 0.042207717323615564,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.1197,
|
|
"mean_token_accuracy": 0.1659408539533615,
|
|
"num_tokens": 1317760.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.304014396667481,
|
|
"epoch": 0.0423230389556473,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.16132398545742035,
|
|
"num_tokens": 1321585.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.135642147064209,
|
|
"epoch": 0.04243836058767904,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.0133,
|
|
"mean_token_accuracy": 0.16338299810886384,
|
|
"num_tokens": 1325132.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.972403478622437,
|
|
"epoch": 0.04255368221971077,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.0241,
|
|
"mean_token_accuracy": 0.17214294523000717,
|
|
"num_tokens": 1328522.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.993261098861694,
|
|
"epoch": 0.04266900385174251,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.9032,
|
|
"mean_token_accuracy": 0.16795059144496918,
|
|
"num_tokens": 1331720.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.163714981079101,
|
|
"epoch": 0.04278432548377425,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.095,
|
|
"mean_token_accuracy": 0.16582898497581483,
|
|
"num_tokens": 1335473.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.191341829299927,
|
|
"epoch": 0.04289964711580598,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.042,
|
|
"mean_token_accuracy": 0.15886857658624648,
|
|
"num_tokens": 1338848.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.23183765411377,
|
|
"epoch": 0.043014968747837716,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.1144,
|
|
"mean_token_accuracy": 0.15058361664414405,
|
|
"num_tokens": 1342260.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.11365852355957,
|
|
"epoch": 0.04313029037986946,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 5.9596,
|
|
"mean_token_accuracy": 0.17216014862060547,
|
|
"num_tokens": 1345864.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.097348260879516,
|
|
"epoch": 0.04324561201190119,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 5.9195,
|
|
"mean_token_accuracy": 0.17186803221702576,
|
|
"num_tokens": 1349787.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.160152959823608,
|
|
"epoch": 0.04336093364393293,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.1034,
|
|
"mean_token_accuracy": 0.1740642584860325,
|
|
"num_tokens": 1353515.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.020594072341919,
|
|
"epoch": 0.043476255275964666,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 5.9085,
|
|
"mean_token_accuracy": 0.17723613381385803,
|
|
"num_tokens": 1356918.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.117224025726318,
|
|
"epoch": 0.0435915769079964,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.0884,
|
|
"mean_token_accuracy": 0.16032198816537857,
|
|
"num_tokens": 1360804.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.18634352684021,
|
|
"epoch": 0.04370689854002814,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.0997,
|
|
"mean_token_accuracy": 0.16843089908361436,
|
|
"num_tokens": 1364246.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.088433122634887,
|
|
"epoch": 0.043822220172059875,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.9804,
|
|
"mean_token_accuracy": 0.17255652248859404,
|
|
"num_tokens": 1367607.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.938327932357788,
|
|
"epoch": 0.04393754180409161,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 5.945,
|
|
"mean_token_accuracy": 0.17490671277046205,
|
|
"num_tokens": 1370886.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.118329954147339,
|
|
"epoch": 0.04405286343612335,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.1318,
|
|
"mean_token_accuracy": 0.16068541407585143,
|
|
"num_tokens": 1375071.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.1548957347869875,
|
|
"epoch": 0.044168185068155084,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.16023668199777602,
|
|
"num_tokens": 1378735.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.100369548797607,
|
|
"epoch": 0.04428350670018682,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.2159,
|
|
"mean_token_accuracy": 0.1529587909579277,
|
|
"num_tokens": 1382397.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.2382121086120605,
|
|
"epoch": 0.04439882833221856,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.0837,
|
|
"mean_token_accuracy": 0.15266251042485238,
|
|
"num_tokens": 1385953.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.095812129974365,
|
|
"epoch": 0.044514149964250294,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.0148,
|
|
"mean_token_accuracy": 0.17738003134727479,
|
|
"num_tokens": 1389621.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.097259616851806,
|
|
"epoch": 0.04462947159628203,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.9071,
|
|
"mean_token_accuracy": 0.18144808411598207,
|
|
"num_tokens": 1392750.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.132451820373535,
|
|
"epoch": 0.04474479322831377,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.0353,
|
|
"mean_token_accuracy": 0.17304997146129608,
|
|
"num_tokens": 1396410.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.112583446502685,
|
|
"epoch": 0.0448601148603455,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.1634,
|
|
"mean_token_accuracy": 0.15278192311525346,
|
|
"num_tokens": 1400340.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.14226188659668,
|
|
"epoch": 0.04497543649237724,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.884,
|
|
"mean_token_accuracy": 0.1774539016187191,
|
|
"num_tokens": 1403841.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.04253191947937,
|
|
"epoch": 0.04509075812440898,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.9487,
|
|
"mean_token_accuracy": 0.1710304409265518,
|
|
"num_tokens": 1407264.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.025435972213745,
|
|
"epoch": 0.04520607975644071,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.085,
|
|
"mean_token_accuracy": 0.15733296647667885,
|
|
"num_tokens": 1411212.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.091107130050659,
|
|
"epoch": 0.04532140138847245,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.9982,
|
|
"mean_token_accuracy": 0.16835324466228485,
|
|
"num_tokens": 1414929.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.109365749359131,
|
|
"epoch": 0.04543672302050419,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.9137,
|
|
"mean_token_accuracy": 0.17214410305023192,
|
|
"num_tokens": 1418211.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.969909429550171,
|
|
"epoch": 0.04555204465253592,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.8435,
|
|
"mean_token_accuracy": 0.1782256096601486,
|
|
"num_tokens": 1421712.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.98684663772583,
|
|
"epoch": 0.04566736628456766,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.9063,
|
|
"mean_token_accuracy": 0.17555052042007446,
|
|
"num_tokens": 1424949.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.9482996463775635,
|
|
"epoch": 0.045782687916599396,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.0216,
|
|
"mean_token_accuracy": 0.17019356042146683,
|
|
"num_tokens": 1428304.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.03603229522705,
|
|
"epoch": 0.04589800954863113,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.9455,
|
|
"mean_token_accuracy": 0.18042387515306474,
|
|
"num_tokens": 1432157.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.0254754543304445,
|
|
"epoch": 0.04601333118066287,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.0697,
|
|
"mean_token_accuracy": 0.15229839086532593,
|
|
"num_tokens": 1435723.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.09818205833435,
|
|
"epoch": 0.046128652812694605,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.886,
|
|
"mean_token_accuracy": 0.16743561923503875,
|
|
"num_tokens": 1439105.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.085743379592896,
|
|
"epoch": 0.04624397444472634,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 6.0632,
|
|
"mean_token_accuracy": 0.1601194992661476,
|
|
"num_tokens": 1442785.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.034141778945923,
|
|
"epoch": 0.04635929607675808,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 5.8992,
|
|
"mean_token_accuracy": 0.18582726866006852,
|
|
"num_tokens": 1446559.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.829406833648681,
|
|
"epoch": 0.046474617708789814,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 5.8353,
|
|
"mean_token_accuracy": 0.18284675031900405,
|
|
"num_tokens": 1449654.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.985281229019165,
|
|
"epoch": 0.04658993934082155,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 5.9864,
|
|
"mean_token_accuracy": 0.16846108585596084,
|
|
"num_tokens": 1453475.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.005276393890381,
|
|
"epoch": 0.04670526097285329,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 5.991,
|
|
"mean_token_accuracy": 0.17077779173851013,
|
|
"num_tokens": 1457088.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.1624187469482425,
|
|
"epoch": 0.046820582604885023,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 5.8905,
|
|
"mean_token_accuracy": 0.17221736386418343,
|
|
"num_tokens": 1460347.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.064001750946045,
|
|
"epoch": 0.04693590423691676,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 5.9405,
|
|
"mean_token_accuracy": 0.17247451096773148,
|
|
"num_tokens": 1463611.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.0143303871154785,
|
|
"epoch": 0.0470512258689485,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 5.9856,
|
|
"mean_token_accuracy": 0.17385794520378112,
|
|
"num_tokens": 1467908.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.1515793800354,
|
|
"epoch": 0.04716654750098023,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 5.9282,
|
|
"mean_token_accuracy": 0.16605019569396973,
|
|
"num_tokens": 1471894.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.052583885192871,
|
|
"epoch": 0.047281869133011974,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 5.9062,
|
|
"mean_token_accuracy": 0.16587038785219194,
|
|
"num_tokens": 1475549.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.853554773330688,
|
|
"epoch": 0.04739719076504371,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 5.896,
|
|
"mean_token_accuracy": 0.16628096103668213,
|
|
"num_tokens": 1478890.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.964755630493164,
|
|
"epoch": 0.04751251239707544,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 5.8508,
|
|
"mean_token_accuracy": 0.17850539237260818,
|
|
"num_tokens": 1482241.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.949054193496704,
|
|
"epoch": 0.04762783402910718,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 5.8955,
|
|
"mean_token_accuracy": 0.1771554946899414,
|
|
"num_tokens": 1486181.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.044420671463013,
|
|
"epoch": 0.04774315566113892,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 6.0385,
|
|
"mean_token_accuracy": 0.1714116469025612,
|
|
"num_tokens": 1489892.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.143317127227784,
|
|
"epoch": 0.04785847729317065,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 5.9539,
|
|
"mean_token_accuracy": 0.16823285222053527,
|
|
"num_tokens": 1493718.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.943177223205566,
|
|
"epoch": 0.04797379892520239,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 5.9111,
|
|
"mean_token_accuracy": 0.16890143752098083,
|
|
"num_tokens": 1497401.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.898208045959473,
|
|
"epoch": 0.048089120557234126,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 5.8469,
|
|
"mean_token_accuracy": 0.17418639361858368,
|
|
"num_tokens": 1500985.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.121327304840088,
|
|
"epoch": 0.04820444218926586,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 5.9324,
|
|
"mean_token_accuracy": 0.17549365013837814,
|
|
"num_tokens": 1505015.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.048160552978516,
|
|
"epoch": 0.0483197638212976,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 5.9068,
|
|
"mean_token_accuracy": 0.16794332191348077,
|
|
"num_tokens": 1508438.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.958954572677612,
|
|
"epoch": 0.048435085453329335,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.0782,
|
|
"mean_token_accuracy": 0.1654080793261528,
|
|
"num_tokens": 1512534.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.037879133224488,
|
|
"epoch": 0.04855040708536107,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 5.8189,
|
|
"mean_token_accuracy": 0.18404786735773088,
|
|
"num_tokens": 1515954.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.936794090270996,
|
|
"epoch": 0.04866572871739281,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.16877272129058837,
|
|
"num_tokens": 1519516.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.045266723632812,
|
|
"epoch": 0.048781050349424544,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 5.9894,
|
|
"mean_token_accuracy": 0.17173408716917038,
|
|
"num_tokens": 1522888.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.019574499130249,
|
|
"epoch": 0.04889637198145628,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 5.8026,
|
|
"mean_token_accuracy": 0.190910604596138,
|
|
"num_tokens": 1526131.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.907725620269775,
|
|
"epoch": 0.04901169361348802,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 5.8298,
|
|
"mean_token_accuracy": 0.18088223934173583,
|
|
"num_tokens": 1529356.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.057435083389282,
|
|
"epoch": 0.04912701524551975,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 6.0237,
|
|
"mean_token_accuracy": 0.1613768979907036,
|
|
"num_tokens": 1533392.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.046886777877807,
|
|
"epoch": 0.049242336877551494,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 6.0219,
|
|
"mean_token_accuracy": 0.16301709562540054,
|
|
"num_tokens": 1536913.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.077149391174316,
|
|
"epoch": 0.04935765850958323,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 5.9527,
|
|
"mean_token_accuracy": 0.17649648189544678,
|
|
"num_tokens": 1540642.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.994723176956176,
|
|
"epoch": 0.04947298014161496,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 5.9701,
|
|
"mean_token_accuracy": 0.1693198561668396,
|
|
"num_tokens": 1544347.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.120733404159546,
|
|
"epoch": 0.049588301773646704,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 6.0676,
|
|
"mean_token_accuracy": 0.15186913609504699,
|
|
"num_tokens": 1548505.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.001276922225952,
|
|
"epoch": 0.04970362340567844,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 5.931,
|
|
"mean_token_accuracy": 0.1802959769964218,
|
|
"num_tokens": 1551687.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.00238070487976,
|
|
"epoch": 0.04981894503771017,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 5.9427,
|
|
"mean_token_accuracy": 0.1708983987569809,
|
|
"num_tokens": 1555148.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.942540502548217,
|
|
"epoch": 0.04993426666974191,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 5.8453,
|
|
"mean_token_accuracy": 0.17310257256031036,
|
|
"num_tokens": 1559009.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.964804410934448,
|
|
"epoch": 0.05004958830177365,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 5.9052,
|
|
"mean_token_accuracy": 0.16347438097000122,
|
|
"num_tokens": 1562494.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.0486002445220945,
|
|
"epoch": 0.05016490993380538,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 5.8162,
|
|
"mean_token_accuracy": 0.17443151026964188,
|
|
"num_tokens": 1566293.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.969220590591431,
|
|
"epoch": 0.05028023156583712,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 6.0235,
|
|
"mean_token_accuracy": 0.1654578074812889,
|
|
"num_tokens": 1570012.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.079716730117798,
|
|
"epoch": 0.050395553197868856,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 5.8509,
|
|
"mean_token_accuracy": 0.17171062529087067,
|
|
"num_tokens": 1573478.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.852660655975342,
|
|
"epoch": 0.05051087482990059,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 5.7515,
|
|
"mean_token_accuracy": 0.18383088037371637,
|
|
"num_tokens": 1576984.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.794715404510498,
|
|
"epoch": 0.05062619646193233,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 5.8732,
|
|
"mean_token_accuracy": 0.1715984523296356,
|
|
"num_tokens": 1580456.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.165300703048706,
|
|
"epoch": 0.050741518093964065,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 6.068,
|
|
"mean_token_accuracy": 0.16014692410826684,
|
|
"num_tokens": 1584478.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.224728631973266,
|
|
"epoch": 0.0508568397259958,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 5.9531,
|
|
"mean_token_accuracy": 0.16508053988218307,
|
|
"num_tokens": 1587803.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.028619384765625,
|
|
"epoch": 0.05097216135802754,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 5.8516,
|
|
"mean_token_accuracy": 0.1806570842862129,
|
|
"num_tokens": 1591191.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.878174543380737,
|
|
"epoch": 0.051087482990059274,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 6.0232,
|
|
"mean_token_accuracy": 0.16577282100915908,
|
|
"num_tokens": 1594452.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.947756052017212,
|
|
"epoch": 0.051202804622091015,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 5.9084,
|
|
"mean_token_accuracy": 0.18042546659708023,
|
|
"num_tokens": 1598006.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.002830839157104,
|
|
"epoch": 0.05131812625412275,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 5.6865,
|
|
"mean_token_accuracy": 0.19325577169656755,
|
|
"num_tokens": 1601472.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.937723159790039,
|
|
"epoch": 0.05143344788615448,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 5.9211,
|
|
"mean_token_accuracy": 0.17626190781593323,
|
|
"num_tokens": 1604894.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.92651252746582,
|
|
"epoch": 0.051548769518186224,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 5.9096,
|
|
"mean_token_accuracy": 0.1714556947350502,
|
|
"num_tokens": 1608179.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.002237796783447,
|
|
"epoch": 0.05166409115021796,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 6.0061,
|
|
"mean_token_accuracy": 0.17283603101968764,
|
|
"num_tokens": 1611758.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.978145456314087,
|
|
"epoch": 0.05177941278224969,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 5.8262,
|
|
"mean_token_accuracy": 0.17325987815856933,
|
|
"num_tokens": 1615164.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.02099723815918,
|
|
"epoch": 0.05189473441428143,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 5.9476,
|
|
"mean_token_accuracy": 0.16971101015806198,
|
|
"num_tokens": 1619090.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.069686317443848,
|
|
"epoch": 0.05201005604631317,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 5.9215,
|
|
"mean_token_accuracy": 0.16354941725730895,
|
|
"num_tokens": 1622631.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.98136305809021,
|
|
"epoch": 0.0521253776783449,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 5.8429,
|
|
"mean_token_accuracy": 0.17021736949682237,
|
|
"num_tokens": 1626006.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.961228561401367,
|
|
"epoch": 0.05224069931037664,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 5.9409,
|
|
"mean_token_accuracy": 0.16348131597042084,
|
|
"num_tokens": 1629645.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.073096370697021,
|
|
"epoch": 0.05235602094240838,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 5.8314,
|
|
"mean_token_accuracy": 0.1716787710785866,
|
|
"num_tokens": 1633368.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.014525318145752,
|
|
"epoch": 0.05247134257444011,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 5.7944,
|
|
"mean_token_accuracy": 0.18000184893608093,
|
|
"num_tokens": 1636918.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.812050247192383,
|
|
"epoch": 0.05258666420647185,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 5.8593,
|
|
"mean_token_accuracy": 0.1746572181582451,
|
|
"num_tokens": 1640128.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.824452114105225,
|
|
"epoch": 0.052701985838503586,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.1883847162127495,
|
|
"num_tokens": 1643335.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.023464822769165,
|
|
"epoch": 0.05281730747053532,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 5.9578,
|
|
"mean_token_accuracy": 0.1665058344602585,
|
|
"num_tokens": 1647175.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.011173248291016,
|
|
"epoch": 0.05293262910256706,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 5.8547,
|
|
"mean_token_accuracy": 0.17507839798927308,
|
|
"num_tokens": 1650561.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.886897420883178,
|
|
"epoch": 0.053047950734598795,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 5.8056,
|
|
"mean_token_accuracy": 0.1859607517719269,
|
|
"num_tokens": 1653829.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.998986721038818,
|
|
"epoch": 0.053163272366630536,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 5.9753,
|
|
"mean_token_accuracy": 0.16849244385957718,
|
|
"num_tokens": 1657573.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.099408769607544,
|
|
"epoch": 0.05327859399866227,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 5.8354,
|
|
"mean_token_accuracy": 0.17568505853414534,
|
|
"num_tokens": 1661467.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.961269044876099,
|
|
"epoch": 0.053393915630694004,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 5.9797,
|
|
"mean_token_accuracy": 0.17760460525751115,
|
|
"num_tokens": 1665306.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.888813829421997,
|
|
"epoch": 0.053509237262725745,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.18595115840435028,
|
|
"num_tokens": 1668926.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.968820381164551,
|
|
"epoch": 0.05362455889475748,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 5.8748,
|
|
"mean_token_accuracy": 0.18151271790266038,
|
|
"num_tokens": 1672620.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.996521663665772,
|
|
"epoch": 0.05373988052678921,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.180991268157959,
|
|
"num_tokens": 1676603.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.947320413589478,
|
|
"epoch": 0.053855202158820954,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 5.7471,
|
|
"mean_token_accuracy": 0.18288988918066024,
|
|
"num_tokens": 1680241.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.833743000030518,
|
|
"epoch": 0.05397052379085269,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 5.7961,
|
|
"mean_token_accuracy": 0.19043108075857162,
|
|
"num_tokens": 1683589.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.9138342380523685,
|
|
"epoch": 0.05408584542288442,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 5.8103,
|
|
"mean_token_accuracy": 0.17725066244602203,
|
|
"num_tokens": 1686648.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.054792833328247,
|
|
"epoch": 0.05420116705491616,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 5.8844,
|
|
"mean_token_accuracy": 0.17706977277994157,
|
|
"num_tokens": 1690343.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.926385879516602,
|
|
"epoch": 0.0543164886869479,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 5.8564,
|
|
"mean_token_accuracy": 0.17736780345439912,
|
|
"num_tokens": 1694078.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.011347770690918,
|
|
"epoch": 0.05443181031897963,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 5.9392,
|
|
"mean_token_accuracy": 0.16562311947345734,
|
|
"num_tokens": 1697818.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.03399772644043,
|
|
"epoch": 0.05454713195101137,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 5.8952,
|
|
"mean_token_accuracy": 0.1669160932302475,
|
|
"num_tokens": 1701178.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.959165763854981,
|
|
"epoch": 0.05466245358304311,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 5.6728,
|
|
"mean_token_accuracy": 0.1890374541282654,
|
|
"num_tokens": 1704405.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.96498966217041,
|
|
"epoch": 0.05477777521507484,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 5.91,
|
|
"mean_token_accuracy": 0.17413416504859924,
|
|
"num_tokens": 1708187.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.916409683227539,
|
|
"epoch": 0.05489309684710658,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 5.836,
|
|
"mean_token_accuracy": 0.17607877999544144,
|
|
"num_tokens": 1711776.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.943102693557739,
|
|
"epoch": 0.055008418479138316,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 5.7583,
|
|
"mean_token_accuracy": 0.17656171321868896,
|
|
"num_tokens": 1715524.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.972788715362549,
|
|
"epoch": 0.05512374011117006,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 5.8485,
|
|
"mean_token_accuracy": 0.17465774863958358,
|
|
"num_tokens": 1718988.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.933341407775879,
|
|
"epoch": 0.05523906174320179,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 5.7456,
|
|
"mean_token_accuracy": 0.1756378412246704,
|
|
"num_tokens": 1722446.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.846967792510986,
|
|
"epoch": 0.055354383375233525,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 5.801,
|
|
"mean_token_accuracy": 0.18985700905323027,
|
|
"num_tokens": 1725950.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.887716150283813,
|
|
"epoch": 0.055469705007265266,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.1676909387111664,
|
|
"num_tokens": 1729486.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.06097640991211,
|
|
"epoch": 0.055585026639297,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 5.7999,
|
|
"mean_token_accuracy": 0.17969767153263091,
|
|
"num_tokens": 1733581.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.961601114273071,
|
|
"epoch": 0.055700348271328734,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 5.8103,
|
|
"mean_token_accuracy": 0.17059036940336228,
|
|
"num_tokens": 1736989.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.871474742889404,
|
|
"epoch": 0.055815669903360475,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 5.7474,
|
|
"mean_token_accuracy": 0.18451255857944487,
|
|
"num_tokens": 1740627.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.97197699546814,
|
|
"epoch": 0.05593099153539221,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 5.9309,
|
|
"mean_token_accuracy": 0.16820127069950103,
|
|
"num_tokens": 1744258.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.9781654357910154,
|
|
"epoch": 0.05604631316742394,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 5.8654,
|
|
"mean_token_accuracy": 0.17026595771312714,
|
|
"num_tokens": 1747735.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.049961376190185,
|
|
"epoch": 0.056161634799455684,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 5.895,
|
|
"mean_token_accuracy": 0.18276432752609253,
|
|
"num_tokens": 1751056.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.875255727767945,
|
|
"epoch": 0.05627695643148742,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.18555193543434143,
|
|
"num_tokens": 1754628.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.923152256011963,
|
|
"epoch": 0.05639227806351915,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 5.9012,
|
|
"mean_token_accuracy": 0.1724686861038208,
|
|
"num_tokens": 1758183.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.082364845275879,
|
|
"epoch": 0.05650759969555089,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 5.9167,
|
|
"mean_token_accuracy": 0.17355914190411567,
|
|
"num_tokens": 1761658.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.140507221221924,
|
|
"epoch": 0.05662292132758263,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 5.8359,
|
|
"mean_token_accuracy": 0.17049692422151566,
|
|
"num_tokens": 1765274.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.967069864273071,
|
|
"epoch": 0.05673824295961436,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 5.9424,
|
|
"mean_token_accuracy": 0.16590988785028457,
|
|
"num_tokens": 1768493.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.900895452499389,
|
|
"epoch": 0.0568535645916461,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 5.7222,
|
|
"mean_token_accuracy": 0.18146858960390091,
|
|
"num_tokens": 1771662.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.986420679092407,
|
|
"epoch": 0.056968886223677837,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 5.8093,
|
|
"mean_token_accuracy": 0.17530354410409926,
|
|
"num_tokens": 1774930.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.018362426757813,
|
|
"epoch": 0.05708420785570957,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 5.9395,
|
|
"mean_token_accuracy": 0.17280193567276,
|
|
"num_tokens": 1778298.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.048731088638306,
|
|
"epoch": 0.05719952948774131,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 5.8415,
|
|
"mean_token_accuracy": 0.18023997843265532,
|
|
"num_tokens": 1781874.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.981462287902832,
|
|
"epoch": 0.057314851119773046,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 5.6777,
|
|
"mean_token_accuracy": 0.2011535122990608,
|
|
"num_tokens": 1785292.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.803164339065551,
|
|
"epoch": 0.05743017275180479,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 5.6443,
|
|
"mean_token_accuracy": 0.19873524010181426,
|
|
"num_tokens": 1788574.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.83751654624939,
|
|
"epoch": 0.05754549438383652,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 5.7009,
|
|
"mean_token_accuracy": 0.19297950863838195,
|
|
"num_tokens": 1792234.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.925842523574829,
|
|
"epoch": 0.057660816015868255,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 5.8405,
|
|
"mean_token_accuracy": 0.1691351890563965,
|
|
"num_tokens": 1795533.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.944417428970337,
|
|
"epoch": 0.057776137647899996,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 5.7873,
|
|
"mean_token_accuracy": 0.17935052365064622,
|
|
"num_tokens": 1798947.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.795767736434937,
|
|
"epoch": 0.05789145927993173,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 5.7801,
|
|
"mean_token_accuracy": 0.18194403350353242,
|
|
"num_tokens": 1802650.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.996693658828735,
|
|
"epoch": 0.058006780911963464,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 5.836,
|
|
"mean_token_accuracy": 0.17214784324169158,
|
|
"num_tokens": 1806065.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.964071035385132,
|
|
"epoch": 0.058122102543995205,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 5.9768,
|
|
"mean_token_accuracy": 0.18485443517565728,
|
|
"num_tokens": 1809882.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.935113096237183,
|
|
"epoch": 0.05823742417602694,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 5.7451,
|
|
"mean_token_accuracy": 0.1786161482334137,
|
|
"num_tokens": 1813648.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.990963125228882,
|
|
"epoch": 0.05835274580805867,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.17053087651729584,
|
|
"num_tokens": 1817553.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.033108186721802,
|
|
"epoch": 0.058468067440090414,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 5.7666,
|
|
"mean_token_accuracy": 0.1837624117732048,
|
|
"num_tokens": 1821078.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.046620416641235,
|
|
"epoch": 0.05858338907212215,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 5.8797,
|
|
"mean_token_accuracy": 0.1781151756644249,
|
|
"num_tokens": 1824896.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.995814037322998,
|
|
"epoch": 0.05869871070415388,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.170192851126194,
|
|
"num_tokens": 1828323.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.9400177001953125,
|
|
"epoch": 0.05881403233618562,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 5.888,
|
|
"mean_token_accuracy": 0.16171274781227113,
|
|
"num_tokens": 1831735.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.985615634918213,
|
|
"epoch": 0.05892935396821736,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 5.9184,
|
|
"mean_token_accuracy": 0.1774922326207161,
|
|
"num_tokens": 1835265.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.959656715393066,
|
|
"epoch": 0.05904467560024909,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 5.8164,
|
|
"mean_token_accuracy": 0.1745815619826317,
|
|
"num_tokens": 1839234.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.024081516265869,
|
|
"epoch": 0.05915999723228083,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 5.9503,
|
|
"mean_token_accuracy": 0.16474941223859788,
|
|
"num_tokens": 1843016.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.953095149993897,
|
|
"epoch": 0.059275318864312566,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 5.7423,
|
|
"mean_token_accuracy": 0.1804298684000969,
|
|
"num_tokens": 1846675.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.886389112472534,
|
|
"epoch": 0.05939064049634431,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 5.8667,
|
|
"mean_token_accuracy": 0.16685420721769334,
|
|
"num_tokens": 1850140.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.942398118972778,
|
|
"epoch": 0.05950596212837604,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 5.8095,
|
|
"mean_token_accuracy": 0.1771612733602524,
|
|
"num_tokens": 1853431.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.950789165496826,
|
|
"epoch": 0.059621283760407776,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 5.7989,
|
|
"mean_token_accuracy": 0.1790459156036377,
|
|
"num_tokens": 1857021.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.885915088653564,
|
|
"epoch": 0.05973660539243952,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 5.8019,
|
|
"mean_token_accuracy": 0.17533093094825744,
|
|
"num_tokens": 1860714.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.949287080764771,
|
|
"epoch": 0.05985192702447125,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 6.042,
|
|
"mean_token_accuracy": 0.1595090538263321,
|
|
"num_tokens": 1865086.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.954693222045899,
|
|
"epoch": 0.059967248656502985,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 5.6983,
|
|
"mean_token_accuracy": 0.19203769713640212,
|
|
"num_tokens": 1868745.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.939291906356812,
|
|
"epoch": 0.060082570288534726,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 5.7347,
|
|
"mean_token_accuracy": 0.18128696829080582,
|
|
"num_tokens": 1872447.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.947452211380005,
|
|
"epoch": 0.06019789192056646,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 5.8386,
|
|
"mean_token_accuracy": 0.1810000628232956,
|
|
"num_tokens": 1876291.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.842586898803711,
|
|
"epoch": 0.060313213552598194,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 5.6949,
|
|
"mean_token_accuracy": 0.1871401250362396,
|
|
"num_tokens": 1879488.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.857621097564698,
|
|
"epoch": 0.060428535184629935,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 5.8766,
|
|
"mean_token_accuracy": 0.17095570862293244,
|
|
"num_tokens": 1883310.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.030225658416748,
|
|
"epoch": 0.06054385681666167,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 5.8159,
|
|
"mean_token_accuracy": 0.17086649388074876,
|
|
"num_tokens": 1886712.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.9798778057098385,
|
|
"epoch": 0.0606591784486934,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 5.8681,
|
|
"mean_token_accuracy": 0.18399405777454375,
|
|
"num_tokens": 1890258.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.999663496017456,
|
|
"epoch": 0.060774500080725144,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 5.8544,
|
|
"mean_token_accuracy": 0.17043397426605225,
|
|
"num_tokens": 1893925.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.943537473678589,
|
|
"epoch": 0.06088982171275688,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 5.7493,
|
|
"mean_token_accuracy": 0.18034369498491287,
|
|
"num_tokens": 1897487.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.919133949279785,
|
|
"epoch": 0.06100514334478861,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 5.7486,
|
|
"mean_token_accuracy": 0.18549261689186097,
|
|
"num_tokens": 1901152.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.887555360794067,
|
|
"epoch": 0.06112046497682035,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.1819697141647339,
|
|
"num_tokens": 1904518.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.953393745422363,
|
|
"epoch": 0.06123578660885209,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.18592287749052047,
|
|
"num_tokens": 1907976.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.946745157241821,
|
|
"epoch": 0.06135110824088383,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 5.7784,
|
|
"mean_token_accuracy": 0.1780816301703453,
|
|
"num_tokens": 1911553.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.9896095275878904,
|
|
"epoch": 0.06146642987291556,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 5.7916,
|
|
"mean_token_accuracy": 0.1862002745270729,
|
|
"num_tokens": 1915462.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.965845441818237,
|
|
"epoch": 0.061581751504947296,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 5.8365,
|
|
"mean_token_accuracy": 0.16558931469917298,
|
|
"num_tokens": 1918894.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.826010036468506,
|
|
"epoch": 0.06169707313697904,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 5.7633,
|
|
"mean_token_accuracy": 0.18768141716718673,
|
|
"num_tokens": 1922258.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.900754165649414,
|
|
"epoch": 0.06181239476901077,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 5.7916,
|
|
"mean_token_accuracy": 0.17888938635587692,
|
|
"num_tokens": 1925996.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.926787519454956,
|
|
"epoch": 0.061927716401042505,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 5.8133,
|
|
"mean_token_accuracy": 0.17930764704942703,
|
|
"num_tokens": 1929567.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.995348215103149,
|
|
"epoch": 0.062043038033074246,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 5.8321,
|
|
"mean_token_accuracy": 0.17503068894147872,
|
|
"num_tokens": 1933300.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.927381324768066,
|
|
"epoch": 0.06215835966510598,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 5.7733,
|
|
"mean_token_accuracy": 0.18090068250894548,
|
|
"num_tokens": 1937003.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.824659395217895,
|
|
"epoch": 0.062273681297137715,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 5.6963,
|
|
"mean_token_accuracy": 0.18889523148536683,
|
|
"num_tokens": 1940846.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.930697011947632,
|
|
"epoch": 0.062389002929169456,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 5.8229,
|
|
"mean_token_accuracy": 0.17768474370241166,
|
|
"num_tokens": 1944197.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.140079402923584,
|
|
"epoch": 0.0625043245612012,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.1695894032716751,
|
|
"num_tokens": 1948108.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.948258304595948,
|
|
"epoch": 0.06261964619323293,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 5.81,
|
|
"mean_token_accuracy": 0.17596178352832795,
|
|
"num_tokens": 1951623.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.806301736831665,
|
|
"epoch": 0.06273496782526466,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 5.7885,
|
|
"mean_token_accuracy": 0.1726483076810837,
|
|
"num_tokens": 1955564.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.905673837661743,
|
|
"epoch": 0.0628502894572964,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 5.8489,
|
|
"mean_token_accuracy": 0.17810622453689576,
|
|
"num_tokens": 1959514.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.030191469192505,
|
|
"epoch": 0.06296561108932813,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 5.8108,
|
|
"mean_token_accuracy": 0.17116886079311372,
|
|
"num_tokens": 1962924.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.898357677459717,
|
|
"epoch": 0.06308093272135987,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 5.6772,
|
|
"mean_token_accuracy": 0.17467646449804305,
|
|
"num_tokens": 1966098.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.976005268096924,
|
|
"epoch": 0.06319625435339161,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 5.8069,
|
|
"mean_token_accuracy": 0.17969678938388825,
|
|
"num_tokens": 1969599.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.94629168510437,
|
|
"epoch": 0.06331157598542335,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 5.8193,
|
|
"mean_token_accuracy": 0.17965982407331466,
|
|
"num_tokens": 1973347.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.924161624908447,
|
|
"epoch": 0.06342689761745508,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 5.7161,
|
|
"mean_token_accuracy": 0.17836985141038894,
|
|
"num_tokens": 1976685.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.856116580963135,
|
|
"epoch": 0.06354221924948682,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 5.826,
|
|
"mean_token_accuracy": 0.1835630714893341,
|
|
"num_tokens": 1980352.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.984286499023438,
|
|
"epoch": 0.06365754088151855,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 5.8972,
|
|
"mean_token_accuracy": 0.1611005485057831,
|
|
"num_tokens": 1984141.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.0155855178833,
|
|
"epoch": 0.06377286251355029,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 5.9249,
|
|
"mean_token_accuracy": 0.1729442611336708,
|
|
"num_tokens": 1987996.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.955809593200684,
|
|
"epoch": 0.06388818414558203,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 5.9013,
|
|
"mean_token_accuracy": 0.17645688205957413,
|
|
"num_tokens": 1991654.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.891233110427857,
|
|
"epoch": 0.06400350577761377,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 5.6333,
|
|
"mean_token_accuracy": 0.19638641327619552,
|
|
"num_tokens": 1995249.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.824556112289429,
|
|
"epoch": 0.0641188274096455,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 5.6957,
|
|
"mean_token_accuracy": 0.1859187066555023,
|
|
"num_tokens": 1998872.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.96528844833374,
|
|
"epoch": 0.06423414904167724,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 5.7449,
|
|
"mean_token_accuracy": 0.1820010259747505,
|
|
"num_tokens": 2002443.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.913536262512207,
|
|
"epoch": 0.06434947067370897,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 5.8393,
|
|
"mean_token_accuracy": 0.17568698674440383,
|
|
"num_tokens": 2006019.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.92832498550415,
|
|
"epoch": 0.06446479230574072,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 5.7788,
|
|
"mean_token_accuracy": 0.18022071868181228,
|
|
"num_tokens": 2009859.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.900950813293457,
|
|
"epoch": 0.06458011393777245,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 5.7762,
|
|
"mean_token_accuracy": 0.17470620572566986,
|
|
"num_tokens": 2013862.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.877919340133667,
|
|
"epoch": 0.06469543556980419,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 5.698,
|
|
"mean_token_accuracy": 0.18921110332012175,
|
|
"num_tokens": 2017526.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.7786976337432865,
|
|
"epoch": 0.06481075720183592,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 5.7208,
|
|
"mean_token_accuracy": 0.18244462758302687,
|
|
"num_tokens": 2021111.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.872343063354492,
|
|
"epoch": 0.06492607883386765,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 5.6845,
|
|
"mean_token_accuracy": 0.18511382043361663,
|
|
"num_tokens": 2024553.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.954916715621948,
|
|
"epoch": 0.06504140046589939,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 5.7107,
|
|
"mean_token_accuracy": 0.1827213019132614,
|
|
"num_tokens": 2027970.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.756761932373047,
|
|
"epoch": 0.06515672209793114,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 5.6702,
|
|
"mean_token_accuracy": 0.20236370116472244,
|
|
"num_tokens": 2031523.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.8211219787597654,
|
|
"epoch": 0.06527204372996287,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 5.8677,
|
|
"mean_token_accuracy": 0.1663561798632145,
|
|
"num_tokens": 2035198.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.059328842163086,
|
|
"epoch": 0.0653873653619946,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 5.9353,
|
|
"mean_token_accuracy": 0.1675690606236458,
|
|
"num_tokens": 2038778.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.051515531539917,
|
|
"epoch": 0.06550268699402634,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 5.8029,
|
|
"mean_token_accuracy": 0.17415003180503846,
|
|
"num_tokens": 2042251.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.995488786697388,
|
|
"epoch": 0.06561800862605807,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 5.8791,
|
|
"mean_token_accuracy": 0.17844734489917755,
|
|
"num_tokens": 2045935.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.9566343307495115,
|
|
"epoch": 0.0657333302580898,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 5.7437,
|
|
"mean_token_accuracy": 0.17331330105662346,
|
|
"num_tokens": 2049516.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.856440591812134,
|
|
"epoch": 0.06584865189012155,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 5.7463,
|
|
"mean_token_accuracy": 0.19445825666189193,
|
|
"num_tokens": 2052830.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.8048499584197994,
|
|
"epoch": 0.06596397352215329,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 5.7006,
|
|
"mean_token_accuracy": 0.1811637058854103,
|
|
"num_tokens": 2056336.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.909080600738525,
|
|
"epoch": 0.06607929515418502,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.17416435331106186,
|
|
"num_tokens": 2059751.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.998892784118652,
|
|
"epoch": 0.06619461678621676,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 5.8553,
|
|
"mean_token_accuracy": 0.17885113656520843,
|
|
"num_tokens": 2062979.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.949467992782592,
|
|
"epoch": 0.06630993841824849,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 5.8315,
|
|
"mean_token_accuracy": 0.17441229820251464,
|
|
"num_tokens": 2066745.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.934280443191528,
|
|
"epoch": 0.06642526005028024,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 5.7666,
|
|
"mean_token_accuracy": 0.17849702388048172,
|
|
"num_tokens": 2070467.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.865775871276855,
|
|
"epoch": 0.06654058168231197,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 5.7864,
|
|
"mean_token_accuracy": 0.1703128144145012,
|
|
"num_tokens": 2073997.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.855584716796875,
|
|
"epoch": 0.0666559033143437,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 5.6243,
|
|
"mean_token_accuracy": 0.18421716541051864,
|
|
"num_tokens": 2077733.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.8581289768219,
|
|
"epoch": 0.06677122494637544,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 5.846,
|
|
"mean_token_accuracy": 0.1686860553920269,
|
|
"num_tokens": 2081732.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.840925407409668,
|
|
"epoch": 0.06688654657840717,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 5.6462,
|
|
"mean_token_accuracy": 0.18950218558311463,
|
|
"num_tokens": 2085248.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.92666187286377,
|
|
"epoch": 0.06700186821043891,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 5.8288,
|
|
"mean_token_accuracy": 0.17843850553035737,
|
|
"num_tokens": 2088525.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.912039947509766,
|
|
"epoch": 0.06711718984247066,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 5.8602,
|
|
"mean_token_accuracy": 0.17369918674230575,
|
|
"num_tokens": 2092162.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.92589750289917,
|
|
"epoch": 0.06723251147450239,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 5.7752,
|
|
"mean_token_accuracy": 0.1798476554453373,
|
|
"num_tokens": 2095715.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.907165145874023,
|
|
"epoch": 0.06734783310653412,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 5.6509,
|
|
"mean_token_accuracy": 0.18663966953754424,
|
|
"num_tokens": 2099234.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.96983962059021,
|
|
"epoch": 0.06746315473856586,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 5.8222,
|
|
"mean_token_accuracy": 0.17739422619342804,
|
|
"num_tokens": 2102815.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.897303581237793,
|
|
"epoch": 0.06757847637059759,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 5.7615,
|
|
"mean_token_accuracy": 0.17696616649627686,
|
|
"num_tokens": 2106330.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.875334548950195,
|
|
"epoch": 0.06769379800262933,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.17329180538654326,
|
|
"num_tokens": 2109995.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.897954416275025,
|
|
"epoch": 0.06780911963466107,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 5.6407,
|
|
"mean_token_accuracy": 0.18700176626443862,
|
|
"num_tokens": 2113454.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.869551944732666,
|
|
"epoch": 0.06792444126669281,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 5.6079,
|
|
"mean_token_accuracy": 0.19167568385601044,
|
|
"num_tokens": 2117278.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.8110098361969,
|
|
"epoch": 0.06803976289872454,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 5.6603,
|
|
"mean_token_accuracy": 0.19450315982103347,
|
|
"num_tokens": 2120867.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.814122200012207,
|
|
"epoch": 0.06815508453075628,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 5.7054,
|
|
"mean_token_accuracy": 0.18472784012556076,
|
|
"num_tokens": 2124407.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.787468814849854,
|
|
"epoch": 0.06827040616278801,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 5.6004,
|
|
"mean_token_accuracy": 0.19235626608133316,
|
|
"num_tokens": 2127846.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.861639308929443,
|
|
"epoch": 0.06838572779481976,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 5.7809,
|
|
"mean_token_accuracy": 0.1853936105966568,
|
|
"num_tokens": 2131445.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.878793239593506,
|
|
"epoch": 0.06850104942685149,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 5.7767,
|
|
"mean_token_accuracy": 0.17848903089761733,
|
|
"num_tokens": 2134881.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.9426216125488285,
|
|
"epoch": 0.06861637105888323,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 5.6981,
|
|
"mean_token_accuracy": 0.18862006217241287,
|
|
"num_tokens": 2138319.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.799541234970093,
|
|
"epoch": 0.06873169269091496,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 5.745,
|
|
"mean_token_accuracy": 0.1744196221232414,
|
|
"num_tokens": 2141814.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.902230501174927,
|
|
"epoch": 0.0688470143229467,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 5.6872,
|
|
"mean_token_accuracy": 0.17609143555164336,
|
|
"num_tokens": 2145505.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.8899266719818115,
|
|
"epoch": 0.06896233595497843,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 5.845,
|
|
"mean_token_accuracy": 0.1673328161239624,
|
|
"num_tokens": 2149248.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.852873516082764,
|
|
"epoch": 0.06907765758701018,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 5.7569,
|
|
"mean_token_accuracy": 0.1847912088036537,
|
|
"num_tokens": 2153091.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.944181680679321,
|
|
"epoch": 0.06919297921904191,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 5.7445,
|
|
"mean_token_accuracy": 0.17777977138757706,
|
|
"num_tokens": 2156379.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.901312685012817,
|
|
"epoch": 0.06930830085107365,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016168477963195382,
|
|
"loss": 5.7432,
|
|
"mean_token_accuracy": 0.1781095489859581,
|
|
"num_tokens": 2159688.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.8753760814666744,
|
|
"epoch": 0.06942362248310538,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001606685517817114,
|
|
"loss": 5.7125,
|
|
"mean_token_accuracy": 0.18335721343755723,
|
|
"num_tokens": 2163399.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.825629138946534,
|
|
"epoch": 0.06953894411513711,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015965545838587592,
|
|
"loss": 5.6658,
|
|
"mean_token_accuracy": 0.18453023731708526,
|
|
"num_tokens": 2166961.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.967210865020752,
|
|
"epoch": 0.06965426574716885,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015864552721891467,
|
|
"loss": 5.8296,
|
|
"mean_token_accuracy": 0.173856483399868,
|
|
"num_tokens": 2170751.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.871779680252075,
|
|
"epoch": 0.0697695873792006,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00015763878596860076,
|
|
"loss": 5.5952,
|
|
"mean_token_accuracy": 0.19168471097946166,
|
|
"num_tokens": 2173995.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.9133158206939695,
|
|
"epoch": 0.06988490901123233,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00015663526223525412,
|
|
"loss": 5.7981,
|
|
"mean_token_accuracy": 0.17257528752088547,
|
|
"num_tokens": 2178020.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.841315031051636,
|
|
"epoch": 0.07000023064326406,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001556349835309848,
|
|
"loss": 5.7033,
|
|
"mean_token_accuracy": 0.18984519094228744,
|
|
"num_tokens": 2181462.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.83582124710083,
|
|
"epoch": 0.0701155522752958,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001546379772789389,
|
|
"loss": 5.7275,
|
|
"mean_token_accuracy": 0.18719817250967025,
|
|
"num_tokens": 2185051.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.882757949829101,
|
|
"epoch": 0.07023087390732753,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015364427081254622,
|
|
"loss": 5.693,
|
|
"mean_token_accuracy": 0.18115523755550383,
|
|
"num_tokens": 2188601.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.970898580551148,
|
|
"epoch": 0.07034619553935928,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015265389137477165,
|
|
"loss": 5.9445,
|
|
"mean_token_accuracy": 0.16347479522228242,
|
|
"num_tokens": 2192503.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.801324415206909,
|
|
"epoch": 0.07046151717139101,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015166686611736786,
|
|
"loss": 5.6889,
|
|
"mean_token_accuracy": 0.1895558938384056,
|
|
"num_tokens": 2196058.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.904062509536743,
|
|
"epoch": 0.07057683880342275,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015068322210013064,
|
|
"loss": 5.6518,
|
|
"mean_token_accuracy": 0.18439579457044603,
|
|
"num_tokens": 2199603.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.898963117599488,
|
|
"epoch": 0.07069216043545448,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001497029862901578,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.18370536714792252,
|
|
"num_tokens": 2203198.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.877163934707641,
|
|
"epoch": 0.07080748206748622,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014872618556110905,
|
|
"loss": 5.5861,
|
|
"mean_token_accuracy": 0.1993960916996002,
|
|
"num_tokens": 2206785.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.924572372436524,
|
|
"epoch": 0.07092280369951795,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014775284669246992,
|
|
"loss": 5.7431,
|
|
"mean_token_accuracy": 0.17719071507453918,
|
|
"num_tokens": 2209981.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.852168130874634,
|
|
"epoch": 0.0710381253315497,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014678299636881716,
|
|
"loss": 5.6498,
|
|
"mean_token_accuracy": 0.19543715119361876,
|
|
"num_tokens": 2213511.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.903236389160156,
|
|
"epoch": 0.07115344696358143,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001458166611790873,
|
|
"loss": 5.7594,
|
|
"mean_token_accuracy": 0.17716941088438035,
|
|
"num_tokens": 2217055.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.836565446853638,
|
|
"epoch": 0.07126876859561317,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014485386761584773,
|
|
"loss": 5.7244,
|
|
"mean_token_accuracy": 0.18442100286483765,
|
|
"num_tokens": 2220707.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.882135915756225,
|
|
"epoch": 0.0713840902276449,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014389464207457042,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.1915835842490196,
|
|
"num_tokens": 2224139.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.970962762832642,
|
|
"epoch": 0.07149941185967663,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014293901085290795,
|
|
"loss": 5.8265,
|
|
"mean_token_accuracy": 0.1745887279510498,
|
|
"num_tokens": 2227543.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.86533555984497,
|
|
"epoch": 0.07161473349170837,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014198700014997307,
|
|
"loss": 5.6955,
|
|
"mean_token_accuracy": 0.18307476490736008,
|
|
"num_tokens": 2230962.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.823045349121093,
|
|
"epoch": 0.07173005512374012,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00014103863606562016,
|
|
"loss": 5.8664,
|
|
"mean_token_accuracy": 0.17263627648353577,
|
|
"num_tokens": 2234300.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.0413025379180905,
|
|
"epoch": 0.07184537675577185,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014009394459972964,
|
|
"loss": 5.8163,
|
|
"mean_token_accuracy": 0.17336128503084183,
|
|
"num_tokens": 2237852.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.9140955924987795,
|
|
"epoch": 0.07196069838780358,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013915295165149513,
|
|
"loss": 5.6823,
|
|
"mean_token_accuracy": 0.18411603271961213,
|
|
"num_tokens": 2241283.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.917401695251465,
|
|
"epoch": 0.07207602001983532,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013821568301871384,
|
|
"loss": 5.6403,
|
|
"mean_token_accuracy": 0.18913399130105973,
|
|
"num_tokens": 2245001.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.812841796875,
|
|
"epoch": 0.07219134165186705,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00013728216439707862,
|
|
"loss": 5.5786,
|
|
"mean_token_accuracy": 0.19669590145349503,
|
|
"num_tokens": 2248795.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.787139415740967,
|
|
"epoch": 0.0723066632838988,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013635242137947419,
|
|
"loss": 5.7365,
|
|
"mean_token_accuracy": 0.1769968032836914,
|
|
"num_tokens": 2252751.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.7433263778686525,
|
|
"epoch": 0.07242198491593053,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013542647945527498,
|
|
"loss": 5.5649,
|
|
"mean_token_accuracy": 0.19809363186359405,
|
|
"num_tokens": 2256135.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.842635250091552,
|
|
"epoch": 0.07253730654796227,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001345043640096465,
|
|
"loss": 5.732,
|
|
"mean_token_accuracy": 0.178202585875988,
|
|
"num_tokens": 2259850.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.778666305541992,
|
|
"epoch": 0.072652628179994,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013358610032284957,
|
|
"loss": 5.7561,
|
|
"mean_token_accuracy": 0.17977271378040313,
|
|
"num_tokens": 2264029.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.971438598632813,
|
|
"epoch": 0.07276794981202574,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000132671713569547,
|
|
"loss": 5.8181,
|
|
"mean_token_accuracy": 0.1793220043182373,
|
|
"num_tokens": 2267891.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.993105745315551,
|
|
"epoch": 0.07288327144405747,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001317612288181136,
|
|
"loss": 5.8273,
|
|
"mean_token_accuracy": 0.17774962782859802,
|
|
"num_tokens": 2271460.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.005527591705322,
|
|
"epoch": 0.07299859307608922,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013085467102994864,
|
|
"loss": 5.7649,
|
|
"mean_token_accuracy": 0.1819105252623558,
|
|
"num_tokens": 2275052.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.919117641448975,
|
|
"epoch": 0.07311391470812095,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012995206505879198,
|
|
"loss": 5.6432,
|
|
"mean_token_accuracy": 0.1917559579014778,
|
|
"num_tokens": 2278529.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.853388977050781,
|
|
"epoch": 0.07322923634015269,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001290534356500421,
|
|
"loss": 5.6141,
|
|
"mean_token_accuracy": 0.19147582054138185,
|
|
"num_tokens": 2282329.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.8208637714385985,
|
|
"epoch": 0.07334455797218442,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012815880744007827,
|
|
"loss": 5.7601,
|
|
"mean_token_accuracy": 0.18827454447746278,
|
|
"num_tokens": 2285908.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.841382646560669,
|
|
"epoch": 0.07345987960421616,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012726820495558483,
|
|
"loss": 5.8541,
|
|
"mean_token_accuracy": 0.1681294858455658,
|
|
"num_tokens": 2289641.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.810229778289795,
|
|
"epoch": 0.07357520123624789,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012638165261287868,
|
|
"loss": 5.6876,
|
|
"mean_token_accuracy": 0.1776748538017273,
|
|
"num_tokens": 2293624.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.98617467880249,
|
|
"epoch": 0.07369052286827964,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001254991747172402,
|
|
"loss": 5.8017,
|
|
"mean_token_accuracy": 0.16751967519521713,
|
|
"num_tokens": 2297153.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.875821161270141,
|
|
"epoch": 0.07380584450031137,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012462079546224662,
|
|
"loss": 5.7404,
|
|
"mean_token_accuracy": 0.17900559604167937,
|
|
"num_tokens": 2300689.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.8457684993743895,
|
|
"epoch": 0.0739211661323431,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012374653892910896,
|
|
"loss": 5.6499,
|
|
"mean_token_accuracy": 0.18321356177330017,
|
|
"num_tokens": 2304058.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.807973480224609,
|
|
"epoch": 0.07403648776437484,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012287642908601166,
|
|
"loss": 5.6152,
|
|
"mean_token_accuracy": 0.1918272152543068,
|
|
"num_tokens": 2307683.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.94158787727356,
|
|
"epoch": 0.07415180939640657,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012201048978745569,
|
|
"loss": 5.7931,
|
|
"mean_token_accuracy": 0.17333043962717057,
|
|
"num_tokens": 2311032.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.8995307922363285,
|
|
"epoch": 0.07426713102843832,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012114874477360427,
|
|
"loss": 5.773,
|
|
"mean_token_accuracy": 0.18127261102199554,
|
|
"num_tokens": 2314928.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.975750732421875,
|
|
"epoch": 0.07438245266047006,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012029121766963236,
|
|
"loss": 5.7595,
|
|
"mean_token_accuracy": 0.1787267029285431,
|
|
"num_tokens": 2318623.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.825032091140747,
|
|
"epoch": 0.07449777429250179,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011943793198507858,
|
|
"loss": 5.5948,
|
|
"mean_token_accuracy": 0.18955389112234117,
|
|
"num_tokens": 2321785.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.83255524635315,
|
|
"epoch": 0.07461309592453352,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011858891111320104,
|
|
"loss": 5.745,
|
|
"mean_token_accuracy": 0.18807391077280045,
|
|
"num_tokens": 2325180.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.909327840805053,
|
|
"epoch": 0.07472841755656526,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001177441783303359,
|
|
"loss": 5.7308,
|
|
"mean_token_accuracy": 0.17989104986190796,
|
|
"num_tokens": 2328493.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.866626930236817,
|
|
"epoch": 0.07484373918859699,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011690375679525896,
|
|
"loss": 5.6566,
|
|
"mean_token_accuracy": 0.1849685102701187,
|
|
"num_tokens": 2332051.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.878978061676025,
|
|
"epoch": 0.07495906082062874,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011606766954855124,
|
|
"loss": 5.782,
|
|
"mean_token_accuracy": 0.17650873214006424,
|
|
"num_tokens": 2335853.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.877120590209961,
|
|
"epoch": 0.07507438245266047,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00011523593951196702,
|
|
"loss": 5.7839,
|
|
"mean_token_accuracy": 0.1787313148379326,
|
|
"num_tokens": 2339505.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.888944005966186,
|
|
"epoch": 0.07518970408469221,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011440858948780523,
|
|
"loss": 5.648,
|
|
"mean_token_accuracy": 0.19025943726301192,
|
|
"num_tokens": 2343276.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.883348655700684,
|
|
"epoch": 0.07530502571672394,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011358564215828484,
|
|
"loss": 5.7072,
|
|
"mean_token_accuracy": 0.17727108150720597,
|
|
"num_tokens": 2346863.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.899157190322876,
|
|
"epoch": 0.07542034734875568,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011276712008492254,
|
|
"loss": 5.6245,
|
|
"mean_token_accuracy": 0.1821736738085747,
|
|
"num_tokens": 2350342.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.860587739944458,
|
|
"epoch": 0.07553566898078741,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011195304570791451,
|
|
"loss": 5.7677,
|
|
"mean_token_accuracy": 0.18490514606237413,
|
|
"num_tokens": 2354193.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.860347557067871,
|
|
"epoch": 0.07565099061281916,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011114344134552094,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.1743358254432678,
|
|
"num_tokens": 2357732.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.8680308818817135,
|
|
"epoch": 0.07576631224485089,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001103383291934545,
|
|
"loss": 5.6986,
|
|
"mean_token_accuracy": 0.18966690301895142,
|
|
"num_tokens": 2361179.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.902762746810913,
|
|
"epoch": 0.07588163387688263,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010953773132427141,
|
|
"loss": 5.7237,
|
|
"mean_token_accuracy": 0.18750899136066437,
|
|
"num_tokens": 2364619.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.867921209335327,
|
|
"epoch": 0.07599695550891436,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010874166968676677,
|
|
"loss": 5.624,
|
|
"mean_token_accuracy": 0.1921158790588379,
|
|
"num_tokens": 2368077.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.808849239349366,
|
|
"epoch": 0.0761122771409461,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010795016610537251,
|
|
"loss": 5.5476,
|
|
"mean_token_accuracy": 0.19577693492174147,
|
|
"num_tokens": 2371553.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.99078483581543,
|
|
"epoch": 0.07622759877297784,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010716324227955904,
|
|
"loss": 5.8753,
|
|
"mean_token_accuracy": 0.178887477517128,
|
|
"num_tokens": 2375170.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.0094873905181885,
|
|
"epoch": 0.07634292040500958,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001063809197832406,
|
|
"loss": 5.9498,
|
|
"mean_token_accuracy": 0.17053980082273484,
|
|
"num_tokens": 2379132.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.838458633422851,
|
|
"epoch": 0.07645824203704131,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00010560322006418368,
|
|
"loss": 5.6421,
|
|
"mean_token_accuracy": 0.19636096060276031,
|
|
"num_tokens": 2382427.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.909769058227539,
|
|
"epoch": 0.07657356366907304,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010483016444341887,
|
|
"loss": 5.6877,
|
|
"mean_token_accuracy": 0.18689777851104736,
|
|
"num_tokens": 2385832.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.835749197006225,
|
|
"epoch": 0.07668888530110478,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010406177411465654,
|
|
"loss": 5.6831,
|
|
"mean_token_accuracy": 0.1866784766316414,
|
|
"num_tokens": 2389386.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.9221714496612545,
|
|
"epoch": 0.07680420693313651,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010329807014370562,
|
|
"loss": 5.9019,
|
|
"mean_token_accuracy": 0.17093419432640075,
|
|
"num_tokens": 2393429.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.808868741989135,
|
|
"epoch": 0.07691952856516826,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010253907346789632,
|
|
"loss": 5.641,
|
|
"mean_token_accuracy": 0.19787009805440903,
|
|
"num_tokens": 2397176.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.8547920227050785,
|
|
"epoch": 0.0770348501972,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010178480489550596,
|
|
"loss": 5.6565,
|
|
"mean_token_accuracy": 0.1898493230342865,
|
|
"num_tokens": 2400590.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.856926774978637,
|
|
"epoch": 0.07715017182923173,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010103528510518836,
|
|
"loss": 5.5885,
|
|
"mean_token_accuracy": 0.2008676841855049,
|
|
"num_tokens": 2404016.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.910517263412475,
|
|
"epoch": 0.07726549346126346,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001002905346454073,
|
|
"loss": 5.8982,
|
|
"mean_token_accuracy": 0.1793844737112522,
|
|
"num_tokens": 2407670.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.881815624237061,
|
|
"epoch": 0.0773808150932952,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.955057393387285e-05,
|
|
"loss": 5.7359,
|
|
"mean_token_accuracy": 0.18224859237670898,
|
|
"num_tokens": 2411252.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.939905309677124,
|
|
"epoch": 0.07749613672532693,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.88154232569816e-05,
|
|
"loss": 5.7252,
|
|
"mean_token_accuracy": 0.1864408567547798,
|
|
"num_tokens": 2414863.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.834432697296142,
|
|
"epoch": 0.07761145835735868,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.808510276926075e-05,
|
|
"loss": 5.7475,
|
|
"mean_token_accuracy": 0.18663297221064568,
|
|
"num_tokens": 2418631.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.776149415969849,
|
|
"epoch": 0.07772677998939041,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.735963249281549e-05,
|
|
"loss": 5.5018,
|
|
"mean_token_accuracy": 0.20407307893037796,
|
|
"num_tokens": 2422160.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.904548692703247,
|
|
"epoch": 0.07784210162142215,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.663903231677974e-05,
|
|
"loss": 5.5926,
|
|
"mean_token_accuracy": 0.18869720548391342,
|
|
"num_tokens": 2425549.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.968348503112793,
|
|
"epoch": 0.07795742325345388,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.592332199677145e-05,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.18033420145511628,
|
|
"num_tokens": 2429066.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.855821180343628,
|
|
"epoch": 0.07807274488548561,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.521252115435061e-05,
|
|
"loss": 5.6166,
|
|
"mean_token_accuracy": 0.18706774413585664,
|
|
"num_tokens": 2432430.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.864228582382202,
|
|
"epoch": 0.07818806651751736,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.450664927648126e-05,
|
|
"loss": 5.7049,
|
|
"mean_token_accuracy": 0.18496773391962051,
|
|
"num_tokens": 2435992.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.8888238906860355,
|
|
"epoch": 0.0783033881495491,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.380572571499758e-05,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.1733245849609375,
|
|
"num_tokens": 2439828.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.873350429534912,
|
|
"epoch": 0.07841870978158083,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.310976968607307e-05,
|
|
"loss": 5.6646,
|
|
"mean_token_accuracy": 0.18091300278902053,
|
|
"num_tokens": 2443180.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.867839813232422,
|
|
"epoch": 0.07853403141361257,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.241880026969381e-05,
|
|
"loss": 5.7011,
|
|
"mean_token_accuracy": 0.18754260540008544,
|
|
"num_tokens": 2446996.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.80164361000061,
|
|
"epoch": 0.0786493530456443,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.173283640913537e-05,
|
|
"loss": 5.7141,
|
|
"mean_token_accuracy": 0.18556749671697617,
|
|
"num_tokens": 2450396.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.808905076980591,
|
|
"epoch": 0.07876467467767603,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.10518969104436e-05,
|
|
"loss": 5.6768,
|
|
"mean_token_accuracy": 0.18256968408823013,
|
|
"num_tokens": 2454046.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.861512899398804,
|
|
"epoch": 0.07887999630970778,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.037600044191868e-05,
|
|
"loss": 5.6816,
|
|
"mean_token_accuracy": 0.19564589262008666,
|
|
"num_tokens": 2457408.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.9170613288879395,
|
|
"epoch": 0.07899531794173952,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.970516553360383e-05,
|
|
"loss": 5.6414,
|
|
"mean_token_accuracy": 0.18897074982523918,
|
|
"num_tokens": 2460614.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.00361738204956,
|
|
"epoch": 0.07911063957377125,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.903941057677692e-05,
|
|
"loss": 5.7419,
|
|
"mean_token_accuracy": 0.18849702477455138,
|
|
"num_tokens": 2464589.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.914666318893433,
|
|
"epoch": 0.07922596120580298,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.837875382344635e-05,
|
|
"loss": 5.71,
|
|
"mean_token_accuracy": 0.18767609298229218,
|
|
"num_tokens": 2468521.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.012236309051514,
|
|
"epoch": 0.07934128283783472,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.772321338585076e-05,
|
|
"loss": 5.8103,
|
|
"mean_token_accuracy": 0.1738918736577034,
|
|
"num_tokens": 2472550.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.865704870223999,
|
|
"epoch": 0.07945660446986645,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.707280723596242e-05,
|
|
"loss": 5.7704,
|
|
"mean_token_accuracy": 0.1887788027524948,
|
|
"num_tokens": 2475837.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.8779661655426025,
|
|
"epoch": 0.0795719261018982,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.64275532049944e-05,
|
|
"loss": 5.6907,
|
|
"mean_token_accuracy": 0.1883298635482788,
|
|
"num_tokens": 2479695.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.860767936706543,
|
|
"epoch": 0.07968724773392993,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.578746898291198e-05,
|
|
"loss": 5.6996,
|
|
"mean_token_accuracy": 0.17127051651477815,
|
|
"num_tokens": 2483265.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.96067247390747,
|
|
"epoch": 0.07980256936596167,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.515257211794742e-05,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.1758652076125145,
|
|
"num_tokens": 2486902.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.874097728729248,
|
|
"epoch": 0.0799178909979934,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.452288001611896e-05,
|
|
"loss": 5.6078,
|
|
"mean_token_accuracy": 0.18402973413467408,
|
|
"num_tokens": 2490588.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.90870213508606,
|
|
"epoch": 0.08003321263002514,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.389840994075379e-05,
|
|
"loss": 5.7007,
|
|
"mean_token_accuracy": 0.19133578538894652,
|
|
"num_tokens": 2494308.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.883487701416016,
|
|
"epoch": 0.08014853426205687,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.327917901201435e-05,
|
|
"loss": 5.6503,
|
|
"mean_token_accuracy": 0.18506948947906493,
|
|
"num_tokens": 2497582.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.913277721405029,
|
|
"epoch": 0.08026385589408862,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.266520420642931e-05,
|
|
"loss": 5.6347,
|
|
"mean_token_accuracy": 0.18075004816055298,
|
|
"num_tokens": 2501104.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.869554805755615,
|
|
"epoch": 0.08037917752612035,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.205650235642828e-05,
|
|
"loss": 5.5905,
|
|
"mean_token_accuracy": 0.18783737421035768,
|
|
"num_tokens": 2504795.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.787777948379516,
|
|
"epoch": 0.08049449915815209,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.145309014987978e-05,
|
|
"loss": 5.6725,
|
|
"mean_token_accuracy": 0.18464811369776726,
|
|
"num_tokens": 2508453.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.839868593215942,
|
|
"epoch": 0.08060982079018382,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.085498412963437e-05,
|
|
"loss": 5.7402,
|
|
"mean_token_accuracy": 0.17645928561687468,
|
|
"num_tokens": 2512540.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.940521240234375,
|
|
"epoch": 0.08072514242221555,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.026220069307078e-05,
|
|
"loss": 5.8577,
|
|
"mean_token_accuracy": 0.17690686583518983,
|
|
"num_tokens": 2516222.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.8679100513458256,
|
|
"epoch": 0.0808404640542473,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.967475609164621e-05,
|
|
"loss": 5.6599,
|
|
"mean_token_accuracy": 0.1920195996761322,
|
|
"num_tokens": 2519775.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.8282781600952145,
|
|
"epoch": 0.08095578568627904,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.909266643045124e-05,
|
|
"loss": 5.6223,
|
|
"mean_token_accuracy": 0.2000498056411743,
|
|
"num_tokens": 2523054.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.84207124710083,
|
|
"epoch": 0.08107110731831077,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.851594766776802e-05,
|
|
"loss": 5.7304,
|
|
"mean_token_accuracy": 0.18813726007938386,
|
|
"num_tokens": 2526742.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.803877973556519,
|
|
"epoch": 0.0811864289503425,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.794461561463265e-05,
|
|
"loss": 5.4945,
|
|
"mean_token_accuracy": 0.1945540875196457,
|
|
"num_tokens": 2530163.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.861265659332275,
|
|
"epoch": 0.08130175058237424,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.7378685934402e-05,
|
|
"loss": 5.7008,
|
|
"mean_token_accuracy": 0.17946666181087495,
|
|
"num_tokens": 2533741.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.840329599380493,
|
|
"epoch": 0.08141707221440597,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.68181741423242e-05,
|
|
"loss": 5.6015,
|
|
"mean_token_accuracy": 0.18979445099830627,
|
|
"num_tokens": 2537938.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.785916519165039,
|
|
"epoch": 0.08153239384643772,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.626309560511313e-05,
|
|
"loss": 5.5319,
|
|
"mean_token_accuracy": 0.1947310373187065,
|
|
"num_tokens": 2541337.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.92472653388977,
|
|
"epoch": 0.08164771547846945,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.571346554052724e-05,
|
|
"loss": 5.6776,
|
|
"mean_token_accuracy": 0.18145399838685988,
|
|
"num_tokens": 2544989.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.900964832305908,
|
|
"epoch": 0.08176303711050119,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.516929901695249e-05,
|
|
"loss": 5.8239,
|
|
"mean_token_accuracy": 0.17422590553760528,
|
|
"num_tokens": 2548999.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.816109943389892,
|
|
"epoch": 0.08187835874253292,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.463061095298893e-05,
|
|
"loss": 5.6092,
|
|
"mean_token_accuracy": 0.19156042933464051,
|
|
"num_tokens": 2552597.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.812936735153198,
|
|
"epoch": 0.08199368037456466,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.409741611704198e-05,
|
|
"loss": 5.7151,
|
|
"mean_token_accuracy": 0.1860814481973648,
|
|
"num_tokens": 2556737.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.8886556148529055,
|
|
"epoch": 0.08210900200659639,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.35697291269174e-05,
|
|
"loss": 5.6804,
|
|
"mean_token_accuracy": 0.18362195566296577,
|
|
"num_tokens": 2560077.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.833680582046509,
|
|
"epoch": 0.08222432363862814,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.304756444942056e-05,
|
|
"loss": 5.6779,
|
|
"mean_token_accuracy": 0.182516647875309,
|
|
"num_tokens": 2563775.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.853801441192627,
|
|
"epoch": 0.08233964527065987,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.253093639995994e-05,
|
|
"loss": 5.6658,
|
|
"mean_token_accuracy": 0.1911642521619797,
|
|
"num_tokens": 2567583.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.854411458969116,
|
|
"epoch": 0.0824549669026916,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.20198591421544e-05,
|
|
"loss": 5.5698,
|
|
"mean_token_accuracy": 0.19344237744808196,
|
|
"num_tokens": 2570848.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.9046735763549805,
|
|
"epoch": 0.08257028853472334,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.151434668744517e-05,
|
|
"loss": 5.6042,
|
|
"mean_token_accuracy": 0.18267150968313217,
|
|
"num_tokens": 2574235.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.778485107421875,
|
|
"epoch": 0.08268561016675507,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.101441289471153e-05,
|
|
"loss": 5.6608,
|
|
"mean_token_accuracy": 0.18832422345876693,
|
|
"num_tokens": 2577686.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.806268405914307,
|
|
"epoch": 0.08280093179878682,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.052007146989098e-05,
|
|
"loss": 5.7302,
|
|
"mean_token_accuracy": 0.18793513476848603,
|
|
"num_tokens": 2581227.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.84239616394043,
|
|
"epoch": 0.08291625343081856,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.003133596560341e-05,
|
|
"loss": 5.6601,
|
|
"mean_token_accuracy": 0.18548210114240646,
|
|
"num_tokens": 2584854.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.797779703140259,
|
|
"epoch": 0.08303157506285029,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.954821978077952e-05,
|
|
"loss": 5.6139,
|
|
"mean_token_accuracy": 0.17625139504671097,
|
|
"num_tokens": 2588628.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.857074356079101,
|
|
"epoch": 0.08314689669488202,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.907073616029356e-05,
|
|
"loss": 5.6637,
|
|
"mean_token_accuracy": 0.18608733862638474,
|
|
"num_tokens": 2592560.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.805485582351684,
|
|
"epoch": 0.08326221832691376,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.85988981946002e-05,
|
|
"loss": 5.5625,
|
|
"mean_token_accuracy": 0.20002525001764299,
|
|
"num_tokens": 2596252.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.881271839141846,
|
|
"epoch": 0.08337753995894549,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 6.813271881937564e-05,
|
|
"loss": 5.5953,
|
|
"mean_token_accuracy": 0.18805267065763473,
|
|
"num_tokens": 2599292.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.86308970451355,
|
|
"epoch": 0.08349286159097724,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.767221081516286e-05,
|
|
"loss": 5.6025,
|
|
"mean_token_accuracy": 0.19107737988233567,
|
|
"num_tokens": 2602823.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.8659721374511715,
|
|
"epoch": 0.08360818322300897,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.72173868070215e-05,
|
|
"loss": 5.8272,
|
|
"mean_token_accuracy": 0.17026104182004928,
|
|
"num_tokens": 2606462.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.757899570465088,
|
|
"epoch": 0.08372350485504071,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.676825926418149e-05,
|
|
"loss": 5.5631,
|
|
"mean_token_accuracy": 0.19940353631973268,
|
|
"num_tokens": 2609625.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.789992237091065,
|
|
"epoch": 0.08383882648707244,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.632484049970122e-05,
|
|
"loss": 5.5694,
|
|
"mean_token_accuracy": 0.1976591616868973,
|
|
"num_tokens": 2612961.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.872446012496948,
|
|
"epoch": 0.08395414811910418,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.588714267013019e-05,
|
|
"loss": 5.7513,
|
|
"mean_token_accuracy": 0.1677234351634979,
|
|
"num_tokens": 2616541.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.840537118911743,
|
|
"epoch": 0.08406946975113591,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.545517777517544e-05,
|
|
"loss": 5.6732,
|
|
"mean_token_accuracy": 0.18060484379529954,
|
|
"num_tokens": 2619837.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.830622959136963,
|
|
"epoch": 0.08418479138316766,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.502895765737281e-05,
|
|
"loss": 5.6945,
|
|
"mean_token_accuracy": 0.18835976123809814,
|
|
"num_tokens": 2623421.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.878254318237305,
|
|
"epoch": 0.0843001130151994,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.460849400176212e-05,
|
|
"loss": 5.7212,
|
|
"mean_token_accuracy": 0.18410128802061082,
|
|
"num_tokens": 2627049.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.939924526214599,
|
|
"epoch": 0.08441543464723113,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.419379833556694e-05,
|
|
"loss": 5.7946,
|
|
"mean_token_accuracy": 0.17868745923042298,
|
|
"num_tokens": 2630996.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.830370712280273,
|
|
"epoch": 0.08453075627926286,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.378488202787835e-05,
|
|
"loss": 5.6975,
|
|
"mean_token_accuracy": 0.18528998047113418,
|
|
"num_tokens": 2634506.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.976336526870727,
|
|
"epoch": 0.0846460779112946,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.33817562893435e-05,
|
|
"loss": 5.807,
|
|
"mean_token_accuracy": 0.1684302270412445,
|
|
"num_tokens": 2637991.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.8713047981262205,
|
|
"epoch": 0.08476139954332634,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.29844321718582e-05,
|
|
"loss": 5.7775,
|
|
"mean_token_accuracy": 0.18358356654644012,
|
|
"num_tokens": 2641343.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.940098428726197,
|
|
"epoch": 0.08487672117535808,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.259292056826383e-05,
|
|
"loss": 5.6886,
|
|
"mean_token_accuracy": 0.18332623988389968,
|
|
"num_tokens": 2644846.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.876470041275025,
|
|
"epoch": 0.08499204280738981,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.220723221204873e-05,
|
|
"loss": 5.5544,
|
|
"mean_token_accuracy": 0.20129456967115403,
|
|
"num_tokens": 2648511.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.922173070907593,
|
|
"epoch": 0.08510736443942155,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.182737767705406e-05,
|
|
"loss": 5.7541,
|
|
"mean_token_accuracy": 0.18271509408950806,
|
|
"num_tokens": 2651964.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.856949949264527,
|
|
"epoch": 0.08522268607145328,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.145336737718375e-05,
|
|
"loss": 5.6706,
|
|
"mean_token_accuracy": 0.18651386797428132,
|
|
"num_tokens": 2655394.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.797024250030518,
|
|
"epoch": 0.08533800770348501,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.10852115661191e-05,
|
|
"loss": 5.5937,
|
|
"mean_token_accuracy": 0.19032420367002487,
|
|
"num_tokens": 2658544.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.8950660705566404,
|
|
"epoch": 0.08545332933551676,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.072292033703766e-05,
|
|
"loss": 5.7721,
|
|
"mean_token_accuracy": 0.1803768664598465,
|
|
"num_tokens": 2662440.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.858198595046997,
|
|
"epoch": 0.0855686509675485,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.036650362233648e-05,
|
|
"loss": 5.6233,
|
|
"mean_token_accuracy": 0.18592633157968522,
|
|
"num_tokens": 2666053.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.942055749893188,
|
|
"epoch": 0.08568397259958023,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.0015971193359824e-05,
|
|
"loss": 5.7054,
|
|
"mean_token_accuracy": 0.17991868853569032,
|
|
"num_tokens": 2669789.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.904725790023804,
|
|
"epoch": 0.08579929423161196,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.9671332660131306e-05,
|
|
"loss": 5.6692,
|
|
"mean_token_accuracy": 0.1755557268857956,
|
|
"num_tokens": 2673241.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.942894315719604,
|
|
"epoch": 0.0859146158636437,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.933259747109042e-05,
|
|
"loss": 5.7184,
|
|
"mean_token_accuracy": 0.18268384486436845,
|
|
"num_tokens": 2677382.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.917337369918823,
|
|
"epoch": 0.08602993749567543,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.899977491283351e-05,
|
|
"loss": 5.6641,
|
|
"mean_token_accuracy": 0.184697625041008,
|
|
"num_tokens": 2681062.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.846370458602905,
|
|
"epoch": 0.08614525912770718,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.867287410985908e-05,
|
|
"loss": 5.669,
|
|
"mean_token_accuracy": 0.19395075142383575,
|
|
"num_tokens": 2684434.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.845784139633179,
|
|
"epoch": 0.08626058075973891,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.835190402431779e-05,
|
|
"loss": 5.702,
|
|
"mean_token_accuracy": 0.19007138013839722,
|
|
"num_tokens": 2687829.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.854014492034912,
|
|
"epoch": 0.08637590239177065,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.803687345576673e-05,
|
|
"loss": 5.76,
|
|
"mean_token_accuracy": 0.17858438193798065,
|
|
"num_tokens": 2691729.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.931945753097534,
|
|
"epoch": 0.08649122402380238,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.7727791040927977e-05,
|
|
"loss": 5.7358,
|
|
"mean_token_accuracy": 0.18099922984838485,
|
|
"num_tokens": 2695440.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.7577649593353275,
|
|
"epoch": 0.08660654565583412,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.742466525345213e-05,
|
|
"loss": 5.5871,
|
|
"mean_token_accuracy": 0.19895114451646806,
|
|
"num_tokens": 2698729.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.896733093261719,
|
|
"epoch": 0.08672186728786586,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.7127504403685775e-05,
|
|
"loss": 5.7956,
|
|
"mean_token_accuracy": 0.17216781079769133,
|
|
"num_tokens": 2702447.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.870556879043579,
|
|
"epoch": 0.0868371889198976,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.6836316638443664e-05,
|
|
"loss": 5.7025,
|
|
"mean_token_accuracy": 0.1874701648950577,
|
|
"num_tokens": 2706118.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.9966882228851315,
|
|
"epoch": 0.08695251055192933,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.655110994078553e-05,
|
|
"loss": 5.7854,
|
|
"mean_token_accuracy": 0.1823234811425209,
|
|
"num_tokens": 2709277.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.886937379837036,
|
|
"epoch": 0.08706783218396107,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.6271892129797056e-05,
|
|
"loss": 5.5978,
|
|
"mean_token_accuracy": 0.19629868268966674,
|
|
"num_tokens": 2712980.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.8487152576446535,
|
|
"epoch": 0.0871831538159928,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.599867086037556e-05,
|
|
"loss": 5.597,
|
|
"mean_token_accuracy": 0.19256910681724548,
|
|
"num_tokens": 2716713.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.865857839584351,
|
|
"epoch": 0.08729847544802453,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.573145362302012e-05,
|
|
"loss": 5.7265,
|
|
"mean_token_accuracy": 0.1843505859375,
|
|
"num_tokens": 2720194.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.845809364318848,
|
|
"epoch": 0.08741379708005628,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.5470247743626404e-05,
|
|
"loss": 5.634,
|
|
"mean_token_accuracy": 0.18462865501642228,
|
|
"num_tokens": 2723805.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.844692325592041,
|
|
"epoch": 0.08752911871208802,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.5215060383285414e-05,
|
|
"loss": 5.7264,
|
|
"mean_token_accuracy": 0.1962260976433754,
|
|
"num_tokens": 2727703.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.868301677703857,
|
|
"epoch": 0.08764444034411975,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.496589853808759e-05,
|
|
"loss": 5.6757,
|
|
"mean_token_accuracy": 0.1872084081172943,
|
|
"num_tokens": 2731266.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.836242008209228,
|
|
"epoch": 0.08775976197615148,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.47227690389308e-05,
|
|
"loss": 5.5683,
|
|
"mean_token_accuracy": 0.18915383368730546,
|
|
"num_tokens": 2734626.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.845393657684326,
|
|
"epoch": 0.08787508360818322,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.448567855133306e-05,
|
|
"loss": 5.6746,
|
|
"mean_token_accuracy": 0.1825521007180214,
|
|
"num_tokens": 2738339.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.823487186431885,
|
|
"epoch": 0.08799040524021495,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.425463357524986e-05,
|
|
"loss": 5.6666,
|
|
"mean_token_accuracy": 0.18648376315832138,
|
|
"num_tokens": 2741923.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.786055564880371,
|
|
"epoch": 0.0881057268722467,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.402964044489591e-05,
|
|
"loss": 5.4801,
|
|
"mean_token_accuracy": 0.1982437402009964,
|
|
"num_tokens": 2745580.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.856432914733887,
|
|
"epoch": 0.08822104850427843,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.381070532857153e-05,
|
|
"loss": 5.7822,
|
|
"mean_token_accuracy": 0.1699580118060112,
|
|
"num_tokens": 2749700.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.808824825286865,
|
|
"epoch": 0.08833637013631017,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.359783422849357e-05,
|
|
"loss": 5.6238,
|
|
"mean_token_accuracy": 0.18773060888051987,
|
|
"num_tokens": 2753562.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.903875255584717,
|
|
"epoch": 0.0884516917683419,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.3391032980630736e-05,
|
|
"loss": 5.6619,
|
|
"mean_token_accuracy": 0.18679060637950898,
|
|
"num_tokens": 2757307.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.881887483596802,
|
|
"epoch": 0.08856701340037364,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.31903072545437e-05,
|
|
"loss": 5.8205,
|
|
"mean_token_accuracy": 0.17016060128808022,
|
|
"num_tokens": 2760847.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.893349552154541,
|
|
"epoch": 0.08868233503240538,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.29956625532297e-05,
|
|
"loss": 5.6638,
|
|
"mean_token_accuracy": 0.18217516094446182,
|
|
"num_tokens": 2764167.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.893620586395263,
|
|
"epoch": 0.08879765666443712,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.280710421297146e-05,
|
|
"loss": 5.7168,
|
|
"mean_token_accuracy": 0.18524640947580337,
|
|
"num_tokens": 2767721.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.892168569564819,
|
|
"epoch": 0.08891297829646885,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.2624637403191165e-05,
|
|
"loss": 5.8034,
|
|
"mean_token_accuracy": 0.1754418432712555,
|
|
"num_tokens": 2771226.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.858520746231079,
|
|
"epoch": 0.08902829992850059,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.2448267126308605e-05,
|
|
"loss": 5.7244,
|
|
"mean_token_accuracy": 0.17887340784072875,
|
|
"num_tokens": 2774577.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.876883172988892,
|
|
"epoch": 0.08914362156053232,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.2277998217603954e-05,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.18104834407567977,
|
|
"num_tokens": 2777917.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.783895635604859,
|
|
"epoch": 0.08925894319256406,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.211383534508541e-05,
|
|
"loss": 5.584,
|
|
"mean_token_accuracy": 0.1899211063981056,
|
|
"num_tokens": 2781497.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.714986324310303,
|
|
"epoch": 0.0893742648245958,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.1955783009361044e-05,
|
|
"loss": 5.4231,
|
|
"mean_token_accuracy": 0.20165151804685594,
|
|
"num_tokens": 2784865.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.836599111557007,
|
|
"epoch": 0.08948958645662754,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.180384554351543e-05,
|
|
"loss": 5.7004,
|
|
"mean_token_accuracy": 0.18591170608997346,
|
|
"num_tokens": 2788392.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.86369891166687,
|
|
"epoch": 0.08960490808865927,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.1658027112990976e-05,
|
|
"loss": 5.6788,
|
|
"mean_token_accuracy": 0.18462732434272766,
|
|
"num_tokens": 2791729.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.805378437042236,
|
|
"epoch": 0.089720229720691,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.151833171547365e-05,
|
|
"loss": 5.6058,
|
|
"mean_token_accuracy": 0.17330496162176132,
|
|
"num_tokens": 2795020.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.921565437316895,
|
|
"epoch": 0.08983555135272274,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.1384763180783274e-05,
|
|
"loss": 5.8713,
|
|
"mean_token_accuracy": 0.17084920555353164,
|
|
"num_tokens": 2798925.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.898939895629883,
|
|
"epoch": 0.08995087298475447,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.125732517076876e-05,
|
|
"loss": 5.7214,
|
|
"mean_token_accuracy": 0.175909686088562,
|
|
"num_tokens": 2802820.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.816446542739868,
|
|
"epoch": 0.09006619461678622,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.113602117920747e-05,
|
|
"loss": 5.6243,
|
|
"mean_token_accuracy": 0.18796323984861374,
|
|
"num_tokens": 2806404.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.842058992385864,
|
|
"epoch": 0.09018151624881796,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.102085453170966e-05,
|
|
"loss": 5.7177,
|
|
"mean_token_accuracy": 0.17969225496053695,
|
|
"num_tokens": 2810018.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.798509788513184,
|
|
"epoch": 0.09029683788084969,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.091182838562709e-05,
|
|
"loss": 5.4735,
|
|
"mean_token_accuracy": 0.204605133831501,
|
|
"num_tokens": 2813552.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.859037303924561,
|
|
"epoch": 0.09041215951288142,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.0808945729966716e-05,
|
|
"loss": 5.6638,
|
|
"mean_token_accuracy": 0.18273267149925232,
|
|
"num_tokens": 2816932.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.926889038085937,
|
|
"epoch": 0.09052748114491316,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.071220938530844e-05,
|
|
"loss": 5.732,
|
|
"mean_token_accuracy": 0.17735587060451508,
|
|
"num_tokens": 2821011.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.78433403968811,
|
|
"epoch": 0.0906428027769449,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.0621622003728026e-05,
|
|
"loss": 5.5948,
|
|
"mean_token_accuracy": 0.19219622015953064,
|
|
"num_tokens": 2824437.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.8929328441619875,
|
|
"epoch": 0.09075812440897664,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.053718606872433e-05,
|
|
"loss": 5.6588,
|
|
"mean_token_accuracy": 0.17977038770914078,
|
|
"num_tokens": 2827993.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.8677451610565186,
|
|
"epoch": 0.09087344604100837,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.0458903895151134e-05,
|
|
"loss": 5.6096,
|
|
"mean_token_accuracy": 0.1907385990023613,
|
|
"num_tokens": 2831540.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.7865440368652346,
|
|
"epoch": 0.09098876767304011,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.038677762915381e-05,
|
|
"loss": 5.5139,
|
|
"mean_token_accuracy": 0.20241701155900954,
|
|
"num_tokens": 2834858.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.90289716720581,
|
|
"epoch": 0.09110408930507184,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.032080924811033e-05,
|
|
"loss": 5.6936,
|
|
"mean_token_accuracy": 0.18548330515623093,
|
|
"num_tokens": 2838769.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.832921934127808,
|
|
"epoch": 0.09121941093710358,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.026100056057718e-05,
|
|
"loss": 5.6934,
|
|
"mean_token_accuracy": 0.1888474404811859,
|
|
"num_tokens": 2842334.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.924541091918945,
|
|
"epoch": 0.09133473256913532,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.0207353206239764e-05,
|
|
"loss": 5.6677,
|
|
"mean_token_accuracy": 0.18344858288764954,
|
|
"num_tokens": 2845987.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.911198091506958,
|
|
"epoch": 0.09145005420116706,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.0159868655867436e-05,
|
|
"loss": 5.7669,
|
|
"mean_token_accuracy": 0.1773296922445297,
|
|
"num_tokens": 2849485.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.8775896549224855,
|
|
"epoch": 0.09156537583319879,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.011854821127305e-05,
|
|
"loss": 5.7652,
|
|
"mean_token_accuracy": 0.18501010090112685,
|
|
"num_tokens": 2852963.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.832471799850464,
|
|
"epoch": 0.09168069746523053,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.008339300527755e-05,
|
|
"loss": 5.5867,
|
|
"mean_token_accuracy": 0.18908259272575378,
|
|
"num_tokens": 2856503.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.80372371673584,
|
|
"epoch": 0.09179601909726226,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.005440400167859e-05,
|
|
"loss": 5.5307,
|
|
"mean_token_accuracy": 0.19805221408605575,
|
|
"num_tokens": 2859705.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.978820514678955,
|
|
"epoch": 0.091911340729294,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.003158199522442e-05,
|
|
"loss": 5.8295,
|
|
"mean_token_accuracy": 0.16185915768146514,
|
|
"num_tokens": 2863393.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.833522844314575,
|
|
"epoch": 0.09202666236132574,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.0014927611591806e-05,
|
|
"loss": 5.6436,
|
|
"mean_token_accuracy": 0.1944509819149971,
|
|
"num_tokens": 2866937.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.850908374786377,
|
|
"epoch": 0.09214198399335748,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.000444130736916e-05,
|
|
"loss": 5.5159,
|
|
"mean_token_accuracy": 0.20287621319293975,
|
|
"num_tokens": 2870464.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.891473436355591,
|
|
"epoch": 0.09225730562538921,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.0000123370043736e-05,
|
|
"loss": 5.6135,
|
|
"mean_token_accuracy": 0.18089342564344407,
|
|
"num_tokens": 2873738.0,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"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": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4418848364544000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|