3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.034596489609520956,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1680181254144000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|