64b67ff7b5
Model: fpadovani/tur-latn-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed10 Source: Original Platform
6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.299895036737142,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.91241865158081,
|
|
"epoch": 0.00049982506122857,
|
|
"grad_norm": 15.25,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.0786,
|
|
"mean_token_accuracy": 0.0003821215010248125,
|
|
"num_tokens": 8821.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.92625322341919,
|
|
"epoch": 0.00099965012245714,
|
|
"grad_norm": 15.3125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.1925,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 18288.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.931429767608643,
|
|
"epoch": 0.00149947518368571,
|
|
"grad_norm": 17.125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.1343,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 28664.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.981450891494751,
|
|
"epoch": 0.00199930024491428,
|
|
"grad_norm": 18.375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.8816,
|
|
"mean_token_accuracy": 0.00011641443707048893,
|
|
"num_tokens": 37807.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.120096206665039,
|
|
"epoch": 0.00249912530614285,
|
|
"grad_norm": 24.875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.594,
|
|
"mean_token_accuracy": 7.418397581204772e-05,
|
|
"num_tokens": 48086.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 5.490445184707641,
|
|
"epoch": 0.00299895036737142,
|
|
"grad_norm": 29.75,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 12.8914,
|
|
"mean_token_accuracy": 0.00022277416428551078,
|
|
"num_tokens": 57427.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.356700134277344,
|
|
"epoch": 0.00349877542859999,
|
|
"grad_norm": 19.125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 11.7365,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 66662.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.28960599899292,
|
|
"epoch": 0.00399860048982856,
|
|
"grad_norm": 3.28125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.813,
|
|
"mean_token_accuracy": 0.0036690699635073543,
|
|
"num_tokens": 77388.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.716344833374023,
|
|
"epoch": 0.00449842555105713,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.6278,
|
|
"mean_token_accuracy": 0.018409140314906834,
|
|
"num_tokens": 86093.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.73198299407959,
|
|
"epoch": 0.0049982506122857,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.5419,
|
|
"mean_token_accuracy": 0.01724100988358259,
|
|
"num_tokens": 96097.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.63021240234375,
|
|
"epoch": 0.00549807567351427,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.3987,
|
|
"mean_token_accuracy": 0.016340368427336216,
|
|
"num_tokens": 106328.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.640047836303712,
|
|
"epoch": 0.00599790073474284,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1971,
|
|
"mean_token_accuracy": 0.019724822044372557,
|
|
"num_tokens": 114822.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.70162353515625,
|
|
"epoch": 0.00649772579597141,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.1508,
|
|
"mean_token_accuracy": 0.019687474332749845,
|
|
"num_tokens": 124984.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.702993392944336,
|
|
"epoch": 0.00699755085719998,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 10.0193,
|
|
"mean_token_accuracy": 0.043946553766727445,
|
|
"num_tokens": 134054.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.67766580581665,
|
|
"epoch": 0.00749737591842855,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.9853,
|
|
"mean_token_accuracy": 0.04433684386312962,
|
|
"num_tokens": 144561.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.662795639038086,
|
|
"epoch": 0.00799720097965712,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.9556,
|
|
"mean_token_accuracy": 0.04047659859061241,
|
|
"num_tokens": 153701.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.647140789031983,
|
|
"epoch": 0.00849702604088569,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.8424,
|
|
"mean_token_accuracy": 0.045346667617559434,
|
|
"num_tokens": 164291.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.631444549560547,
|
|
"epoch": 0.00899685110211426,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.8041,
|
|
"mean_token_accuracy": 0.048126301541924474,
|
|
"num_tokens": 174136.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.60820140838623,
|
|
"epoch": 0.00949667616334283,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.7185,
|
|
"mean_token_accuracy": 0.05448371171951294,
|
|
"num_tokens": 184368.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.577119731903077,
|
|
"epoch": 0.0099965012245714,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.6957,
|
|
"mean_token_accuracy": 0.053209440782666206,
|
|
"num_tokens": 194659.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.547229385375976,
|
|
"epoch": 0.01049632628579997,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.5748,
|
|
"mean_token_accuracy": 0.050935184210538866,
|
|
"num_tokens": 204875.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.515561866760255,
|
|
"epoch": 0.01099615134702854,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.5761,
|
|
"mean_token_accuracy": 0.05177152454853058,
|
|
"num_tokens": 217033.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.471278762817382,
|
|
"epoch": 0.01149597640825711,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.3897,
|
|
"mean_token_accuracy": 0.05110491365194321,
|
|
"num_tokens": 225946.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.367455768585206,
|
|
"epoch": 0.01199580146948568,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.3339,
|
|
"mean_token_accuracy": 0.047009650617837906,
|
|
"num_tokens": 235413.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.327684879302979,
|
|
"epoch": 0.01249562653071425,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.3556,
|
|
"mean_token_accuracy": 0.041035909205675125,
|
|
"num_tokens": 247913.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.27382516860962,
|
|
"epoch": 0.01299545159194282,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.2148,
|
|
"mean_token_accuracy": 0.05273361578583717,
|
|
"num_tokens": 258486.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.201590824127198,
|
|
"epoch": 0.01349527665317139,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.1446,
|
|
"mean_token_accuracy": 0.052062534168362616,
|
|
"num_tokens": 268027.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.166050720214844,
|
|
"epoch": 0.01399510171439996,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 9.0516,
|
|
"mean_token_accuracy": 0.058280370011925695,
|
|
"num_tokens": 277470.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.048835277557373,
|
|
"epoch": 0.01449492677562853,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.9804,
|
|
"mean_token_accuracy": 0.055930846557021144,
|
|
"num_tokens": 286891.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.00039358139038,
|
|
"epoch": 0.0149947518368571,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.8799,
|
|
"mean_token_accuracy": 0.06509376168251038,
|
|
"num_tokens": 295636.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.860487079620361,
|
|
"epoch": 0.01549457689808567,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.8335,
|
|
"mean_token_accuracy": 0.06008245907723904,
|
|
"num_tokens": 306542.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.832607650756836,
|
|
"epoch": 0.01599440195931424,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.7586,
|
|
"mean_token_accuracy": 0.055786200985312465,
|
|
"num_tokens": 315956.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.66825180053711,
|
|
"epoch": 0.01649422702054281,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.7179,
|
|
"mean_token_accuracy": 0.058467098698019984,
|
|
"num_tokens": 325500.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.602266311645508,
|
|
"epoch": 0.01699405208177138,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.694,
|
|
"mean_token_accuracy": 0.05908621363341808,
|
|
"num_tokens": 335194.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.40921688079834,
|
|
"epoch": 0.01749387714299995,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.6446,
|
|
"mean_token_accuracy": 0.0567587573081255,
|
|
"num_tokens": 344717.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.355875492095947,
|
|
"epoch": 0.01799370220422852,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.5415,
|
|
"mean_token_accuracy": 0.058611446246504786,
|
|
"num_tokens": 354476.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.278519535064698,
|
|
"epoch": 0.01849352726545709,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.583,
|
|
"mean_token_accuracy": 0.05657804422080517,
|
|
"num_tokens": 363939.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.173566627502442,
|
|
"epoch": 0.01899335232668566,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.6021,
|
|
"mean_token_accuracy": 0.056013180688023564,
|
|
"num_tokens": 373187.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 9.077548694610595,
|
|
"epoch": 0.01949317738791423,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.5318,
|
|
"mean_token_accuracy": 0.057455115392804144,
|
|
"num_tokens": 382999.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 9.053006076812744,
|
|
"epoch": 0.0199930024491428,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.5703,
|
|
"mean_token_accuracy": 0.0567587573081255,
|
|
"num_tokens": 392639.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 9.040444278717041,
|
|
"epoch": 0.02049282751037137,
|
|
"grad_norm": 0.66015625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.5588,
|
|
"mean_token_accuracy": 0.05599666722118855,
|
|
"num_tokens": 403089.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.963880825042725,
|
|
"epoch": 0.02099265257159994,
|
|
"grad_norm": 0.6875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.5074,
|
|
"mean_token_accuracy": 0.057358425855636594,
|
|
"num_tokens": 412321.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.893467998504638,
|
|
"epoch": 0.02149247763282851,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.5434,
|
|
"mean_token_accuracy": 0.05354794263839722,
|
|
"num_tokens": 421599.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.862988948822021,
|
|
"epoch": 0.02199230269405708,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.5194,
|
|
"mean_token_accuracy": 0.0537799846380949,
|
|
"num_tokens": 432012.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.87530450820923,
|
|
"epoch": 0.02249212775528565,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.4841,
|
|
"mean_token_accuracy": 0.06368993148207665,
|
|
"num_tokens": 441989.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.893553352355957,
|
|
"epoch": 0.02299195281651422,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.4783,
|
|
"mean_token_accuracy": 0.058648154512047766,
|
|
"num_tokens": 451532.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.84007387161255,
|
|
"epoch": 0.02349177787774279,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.4902,
|
|
"mean_token_accuracy": 0.05833564437925816,
|
|
"num_tokens": 461815.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.851331996917725,
|
|
"epoch": 0.02399160293897136,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.5479,
|
|
"mean_token_accuracy": 0.05575990378856659,
|
|
"num_tokens": 470860.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.771306705474853,
|
|
"epoch": 0.02449142800019993,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.517,
|
|
"mean_token_accuracy": 0.05667851753532886,
|
|
"num_tokens": 480649.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.83043212890625,
|
|
"epoch": 0.0249912530614285,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.4491,
|
|
"mean_token_accuracy": 0.06127909794449806,
|
|
"num_tokens": 489969.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.718154525756836,
|
|
"epoch": 0.02549107812265707,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.4681,
|
|
"mean_token_accuracy": 0.05937696918845177,
|
|
"num_tokens": 500000.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.808709049224854,
|
|
"epoch": 0.02599090318388564,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.4517,
|
|
"mean_token_accuracy": 0.06048737429082394,
|
|
"num_tokens": 510013.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.73866376876831,
|
|
"epoch": 0.026490728245114212,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.4782,
|
|
"mean_token_accuracy": 0.05986591465771198,
|
|
"num_tokens": 519698.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.757024574279786,
|
|
"epoch": 0.02699055330634278,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.4976,
|
|
"mean_token_accuracy": 0.05905666761100292,
|
|
"num_tokens": 531631.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.725329685211182,
|
|
"epoch": 0.02749037836757135,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.4571,
|
|
"mean_token_accuracy": 0.06441094093024731,
|
|
"num_tokens": 542204.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.754269123077393,
|
|
"epoch": 0.02799020342879992,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.383,
|
|
"mean_token_accuracy": 0.05882711634039879,
|
|
"num_tokens": 552269.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.679509162902832,
|
|
"epoch": 0.02849002849002849,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.3972,
|
|
"mean_token_accuracy": 0.05945455729961395,
|
|
"num_tokens": 561907.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.714503383636474,
|
|
"epoch": 0.02898985355125706,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.4968,
|
|
"mean_token_accuracy": 0.06066646538674832,
|
|
"num_tokens": 571283.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.684092807769776,
|
|
"epoch": 0.02948967861248563,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.3882,
|
|
"mean_token_accuracy": 0.06541323848068714,
|
|
"num_tokens": 579896.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.677871990203858,
|
|
"epoch": 0.0299895036737142,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.3593,
|
|
"mean_token_accuracy": 0.07032604105770587,
|
|
"num_tokens": 588856.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.640351676940918,
|
|
"epoch": 0.03048932873494277,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.3892,
|
|
"mean_token_accuracy": 0.06575401201844215,
|
|
"num_tokens": 598910.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.718158340454101,
|
|
"epoch": 0.03098915379617134,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.4433,
|
|
"mean_token_accuracy": 0.062070992588996884,
|
|
"num_tokens": 609248.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.671941089630128,
|
|
"epoch": 0.03148897885739991,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.4592,
|
|
"mean_token_accuracy": 0.06366059742867947,
|
|
"num_tokens": 619352.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.706588459014892,
|
|
"epoch": 0.03198880391862848,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.4661,
|
|
"mean_token_accuracy": 0.06272411942481995,
|
|
"num_tokens": 628300.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.747008895874023,
|
|
"epoch": 0.03248862897985705,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.4018,
|
|
"mean_token_accuracy": 0.06640003323554992,
|
|
"num_tokens": 637969.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.643924903869628,
|
|
"epoch": 0.03298845404108562,
|
|
"grad_norm": 0.70703125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.3336,
|
|
"mean_token_accuracy": 0.0662142876535654,
|
|
"num_tokens": 647518.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.71212978363037,
|
|
"epoch": 0.03348827910231419,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.4995,
|
|
"mean_token_accuracy": 0.06047630980610848,
|
|
"num_tokens": 658292.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.704450511932373,
|
|
"epoch": 0.03398810416354276,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.3777,
|
|
"mean_token_accuracy": 0.06684980280697346,
|
|
"num_tokens": 667872.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.679888439178466,
|
|
"epoch": 0.03448792922477133,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.2947,
|
|
"mean_token_accuracy": 0.0693135604262352,
|
|
"num_tokens": 676976.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.639434337615967,
|
|
"epoch": 0.0349877542859999,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.3251,
|
|
"mean_token_accuracy": 0.06586846001446248,
|
|
"num_tokens": 686149.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.610470199584961,
|
|
"epoch": 0.03548757934722847,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.2287,
|
|
"mean_token_accuracy": 0.07368465512990952,
|
|
"num_tokens": 696439.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.60836582183838,
|
|
"epoch": 0.03598740440845704,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.2631,
|
|
"mean_token_accuracy": 0.07091691121459007,
|
|
"num_tokens": 705902.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.526141834259032,
|
|
"epoch": 0.03648722946968561,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.4111,
|
|
"mean_token_accuracy": 0.06636230871081353,
|
|
"num_tokens": 715973.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.639653873443603,
|
|
"epoch": 0.03698705453091418,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.3087,
|
|
"mean_token_accuracy": 0.06943082325160503,
|
|
"num_tokens": 725463.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.648558807373046,
|
|
"epoch": 0.03748687959214275,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.3425,
|
|
"mean_token_accuracy": 0.06693257950246334,
|
|
"num_tokens": 734941.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.635571098327636,
|
|
"epoch": 0.03798670465337132,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.3602,
|
|
"mean_token_accuracy": 0.06358583979308605,
|
|
"num_tokens": 744302.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.627023029327393,
|
|
"epoch": 0.03848652971459989,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.2859,
|
|
"mean_token_accuracy": 0.07252865135669709,
|
|
"num_tokens": 754341.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.77175464630127,
|
|
"epoch": 0.03898635477582846,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.4077,
|
|
"mean_token_accuracy": 0.06557657048106194,
|
|
"num_tokens": 764087.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.606595134735107,
|
|
"epoch": 0.03948617983705703,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.3433,
|
|
"mean_token_accuracy": 0.06623501852154731,
|
|
"num_tokens": 773728.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.625543308258056,
|
|
"epoch": 0.0399860048982856,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.3526,
|
|
"mean_token_accuracy": 0.06745064929127693,
|
|
"num_tokens": 784839.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.703261470794677,
|
|
"epoch": 0.04048582995951417,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.3402,
|
|
"mean_token_accuracy": 0.07067949250340462,
|
|
"num_tokens": 794514.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.594697570800781,
|
|
"epoch": 0.04098565502074274,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.3258,
|
|
"mean_token_accuracy": 0.07445940934121609,
|
|
"num_tokens": 805254.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.560181427001954,
|
|
"epoch": 0.04148548008197131,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.2289,
|
|
"mean_token_accuracy": 0.07385558374226094,
|
|
"num_tokens": 814135.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.59696340560913,
|
|
"epoch": 0.04198530514319988,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.2591,
|
|
"mean_token_accuracy": 0.07281342111527919,
|
|
"num_tokens": 823531.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.521518802642822,
|
|
"epoch": 0.04248513020442845,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.1938,
|
|
"mean_token_accuracy": 0.073985905200243,
|
|
"num_tokens": 832833.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.501078224182129,
|
|
"epoch": 0.04298495526565702,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.189,
|
|
"mean_token_accuracy": 0.07162816822528839,
|
|
"num_tokens": 841541.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.522215366363525,
|
|
"epoch": 0.04348478032688559,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.2042,
|
|
"mean_token_accuracy": 0.07523961365222931,
|
|
"num_tokens": 850689.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.53344202041626,
|
|
"epoch": 0.04398460538811416,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.1527,
|
|
"mean_token_accuracy": 0.08227219060063362,
|
|
"num_tokens": 860573.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.544331645965576,
|
|
"epoch": 0.04448443044934273,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.2541,
|
|
"mean_token_accuracy": 0.07364307269454003,
|
|
"num_tokens": 869784.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.543445301055907,
|
|
"epoch": 0.0449842555105713,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.216,
|
|
"mean_token_accuracy": 0.07438447847962379,
|
|
"num_tokens": 878716.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.488540935516358,
|
|
"epoch": 0.04548408057179987,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.2861,
|
|
"mean_token_accuracy": 0.07369314283132553,
|
|
"num_tokens": 888334.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.54775972366333,
|
|
"epoch": 0.04598390563302844,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.2092,
|
|
"mean_token_accuracy": 0.0723773218691349,
|
|
"num_tokens": 897701.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.553479957580567,
|
|
"epoch": 0.04648373069425701,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.1875,
|
|
"mean_token_accuracy": 0.07418999038636684,
|
|
"num_tokens": 907278.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.54272518157959,
|
|
"epoch": 0.04698355575548558,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.2054,
|
|
"mean_token_accuracy": 0.07134300135076047,
|
|
"num_tokens": 916194.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.520616245269775,
|
|
"epoch": 0.04748338081671415,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.207,
|
|
"mean_token_accuracy": 0.07744368985295295,
|
|
"num_tokens": 926743.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.551657390594482,
|
|
"epoch": 0.04798320587794272,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.2412,
|
|
"mean_token_accuracy": 0.07283293902873993,
|
|
"num_tokens": 936059.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.39524564743042,
|
|
"epoch": 0.04848303093917129,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.1493,
|
|
"mean_token_accuracy": 0.07938387766480445,
|
|
"num_tokens": 945990.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.566281795501709,
|
|
"epoch": 0.04898285600039986,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.2333,
|
|
"mean_token_accuracy": 0.07900915294885635,
|
|
"num_tokens": 956626.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.450263595581054,
|
|
"epoch": 0.04948268106162843,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 8.179,
|
|
"mean_token_accuracy": 0.07676386944949627,
|
|
"num_tokens": 966200.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.555145835876464,
|
|
"epoch": 0.049982506122857,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 8.1396,
|
|
"mean_token_accuracy": 0.07628392353653908,
|
|
"num_tokens": 975651.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.38605089187622,
|
|
"epoch": 0.05048233118408557,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 8.1242,
|
|
"mean_token_accuracy": 0.08255173414945602,
|
|
"num_tokens": 985264.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.503427410125733,
|
|
"epoch": 0.05098215624531414,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 8.1561,
|
|
"mean_token_accuracy": 0.07891211658716202,
|
|
"num_tokens": 996723.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.348312091827392,
|
|
"epoch": 0.05148198130654271,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 8.1596,
|
|
"mean_token_accuracy": 0.07888874784111977,
|
|
"num_tokens": 1006098.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.483654975891113,
|
|
"epoch": 0.05198180636777128,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 8.1879,
|
|
"mean_token_accuracy": 0.07833293974399566,
|
|
"num_tokens": 1016595.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.348458003997802,
|
|
"epoch": 0.05248163142899985,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000262,
|
|
"loss": 8.0453,
|
|
"mean_token_accuracy": 0.08186679258942604,
|
|
"num_tokens": 1026281.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.519502449035645,
|
|
"epoch": 0.052981456490228424,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 8.146,
|
|
"mean_token_accuracy": 0.083558601886034,
|
|
"num_tokens": 1036697.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.291908359527588,
|
|
"epoch": 0.05348128155145699,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 8.1041,
|
|
"mean_token_accuracy": 0.07858130112290382,
|
|
"num_tokens": 1047433.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.540676879882813,
|
|
"epoch": 0.05398110661268556,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 8.1544,
|
|
"mean_token_accuracy": 0.0775446780025959,
|
|
"num_tokens": 1056774.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.399338245391846,
|
|
"epoch": 0.05448093167391413,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 8.0729,
|
|
"mean_token_accuracy": 0.0837493509054184,
|
|
"num_tokens": 1068056.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.35708408355713,
|
|
"epoch": 0.0549807567351427,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 8.0637,
|
|
"mean_token_accuracy": 0.08658572286367416,
|
|
"num_tokens": 1077698.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.371122455596923,
|
|
"epoch": 0.05548058179637127,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 8.024,
|
|
"mean_token_accuracy": 0.08619759157299996,
|
|
"num_tokens": 1086485.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.344655513763428,
|
|
"epoch": 0.05598040685759984,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 8.0981,
|
|
"mean_token_accuracy": 0.07644926160573959,
|
|
"num_tokens": 1097329.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.330466938018798,
|
|
"epoch": 0.05648023191882841,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 8.0422,
|
|
"mean_token_accuracy": 0.08308784440159797,
|
|
"num_tokens": 1107990.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.35728998184204,
|
|
"epoch": 0.05698005698005698,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.9942,
|
|
"mean_token_accuracy": 0.09085135459899903,
|
|
"num_tokens": 1117798.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.300171279907227,
|
|
"epoch": 0.05747988204128555,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 8.1129,
|
|
"mean_token_accuracy": 0.08207414373755455,
|
|
"num_tokens": 1126976.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.369365787506103,
|
|
"epoch": 0.05797970710251412,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 8.0323,
|
|
"mean_token_accuracy": 0.08670488931238651,
|
|
"num_tokens": 1136410.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.333628749847412,
|
|
"epoch": 0.05847953216374269,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 8.0802,
|
|
"mean_token_accuracy": 0.08314882963895798,
|
|
"num_tokens": 1147162.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.32247953414917,
|
|
"epoch": 0.05897935722497126,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 8.0381,
|
|
"mean_token_accuracy": 0.08616725653409958,
|
|
"num_tokens": 1157578.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.34740982055664,
|
|
"epoch": 0.05947918228619983,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.951,
|
|
"mean_token_accuracy": 0.09069004505872727,
|
|
"num_tokens": 1166914.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.420492267608642,
|
|
"epoch": 0.0599790073474284,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 8.1052,
|
|
"mean_token_accuracy": 0.07346431724727154,
|
|
"num_tokens": 1176740.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.245788764953613,
|
|
"epoch": 0.06047883240865697,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 8.0109,
|
|
"mean_token_accuracy": 0.08016561791300773,
|
|
"num_tokens": 1185497.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.463874912261963,
|
|
"epoch": 0.06097865746988554,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 8.1356,
|
|
"mean_token_accuracy": 0.07826456874608993,
|
|
"num_tokens": 1197390.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.216155338287354,
|
|
"epoch": 0.06147848253111411,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 8.0019,
|
|
"mean_token_accuracy": 0.08540406599640846,
|
|
"num_tokens": 1208293.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.2839524269104,
|
|
"epoch": 0.06197830759234268,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.9785,
|
|
"mean_token_accuracy": 0.08736005201935768,
|
|
"num_tokens": 1217860.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.30297269821167,
|
|
"epoch": 0.06247813265357125,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.9677,
|
|
"mean_token_accuracy": 0.0873529352247715,
|
|
"num_tokens": 1227259.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.173470211029052,
|
|
"epoch": 0.06297795771479982,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.8852,
|
|
"mean_token_accuracy": 0.09442147463560105,
|
|
"num_tokens": 1236775.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.359477519989014,
|
|
"epoch": 0.06347778277602839,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 8.016,
|
|
"mean_token_accuracy": 0.08664593175053596,
|
|
"num_tokens": 1246307.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.253477573394775,
|
|
"epoch": 0.06397760783725696,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.9653,
|
|
"mean_token_accuracy": 0.0844787299633026,
|
|
"num_tokens": 1255197.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.19736557006836,
|
|
"epoch": 0.06447743289848552,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.9582,
|
|
"mean_token_accuracy": 0.08905794695019723,
|
|
"num_tokens": 1265056.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.316370296478272,
|
|
"epoch": 0.0649772579597141,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.9362,
|
|
"mean_token_accuracy": 0.0887321688234806,
|
|
"num_tokens": 1274718.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.240344905853272,
|
|
"epoch": 0.06547708302094267,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 8.0083,
|
|
"mean_token_accuracy": 0.08291322365403175,
|
|
"num_tokens": 1284364.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.259224700927735,
|
|
"epoch": 0.06597690808217124,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.9941,
|
|
"mean_token_accuracy": 0.08638291731476784,
|
|
"num_tokens": 1294186.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.232594108581543,
|
|
"epoch": 0.06647673314339982,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.9088,
|
|
"mean_token_accuracy": 0.08878451958298683,
|
|
"num_tokens": 1303393.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.080247354507446,
|
|
"epoch": 0.06697655820462838,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.9236,
|
|
"mean_token_accuracy": 0.08766417391598225,
|
|
"num_tokens": 1312819.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 8.31481409072876,
|
|
"epoch": 0.06747638326585695,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.9491,
|
|
"mean_token_accuracy": 0.08286305814981461,
|
|
"num_tokens": 1323001.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 8.191842365264893,
|
|
"epoch": 0.06797620832708552,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.9247,
|
|
"mean_token_accuracy": 0.0884876474738121,
|
|
"num_tokens": 1333260.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 8.234783840179443,
|
|
"epoch": 0.0684760333883141,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.9158,
|
|
"mean_token_accuracy": 0.09147168770432472,
|
|
"num_tokens": 1342696.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.155243015289306,
|
|
"epoch": 0.06897585844954265,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.8796,
|
|
"mean_token_accuracy": 0.09066492915153504,
|
|
"num_tokens": 1353104.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 8.149154233932496,
|
|
"epoch": 0.06947568351077123,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.94,
|
|
"mean_token_accuracy": 0.08528364300727845,
|
|
"num_tokens": 1362871.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 8.276353740692139,
|
|
"epoch": 0.0699755085719998,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.9276,
|
|
"mean_token_accuracy": 0.0848144493997097,
|
|
"num_tokens": 1373622.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.046894645690918,
|
|
"epoch": 0.07047533363322837,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.8097,
|
|
"mean_token_accuracy": 0.0927486002445221,
|
|
"num_tokens": 1382707.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 8.174900817871094,
|
|
"epoch": 0.07097515869445695,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.8886,
|
|
"mean_token_accuracy": 0.08790072128176689,
|
|
"num_tokens": 1392048.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 8.099133968353271,
|
|
"epoch": 0.0714749837556855,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.7641,
|
|
"mean_token_accuracy": 0.09442322701215744,
|
|
"num_tokens": 1401387.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 8.091927433013916,
|
|
"epoch": 0.07197480881691408,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.8821,
|
|
"mean_token_accuracy": 0.08691308051347732,
|
|
"num_tokens": 1411404.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 8.135170125961304,
|
|
"epoch": 0.07247463387814265,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.8546,
|
|
"mean_token_accuracy": 0.08999823704361916,
|
|
"num_tokens": 1420292.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 8.09806661605835,
|
|
"epoch": 0.07297445893937123,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.8309,
|
|
"mean_token_accuracy": 0.09208463206887245,
|
|
"num_tokens": 1429990.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 8.13855881690979,
|
|
"epoch": 0.07347428400059978,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.8628,
|
|
"mean_token_accuracy": 0.08753219097852707,
|
|
"num_tokens": 1440148.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 8.03170199394226,
|
|
"epoch": 0.07397410906182836,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.7852,
|
|
"mean_token_accuracy": 0.08861454948782921,
|
|
"num_tokens": 1450015.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 8.133480024337768,
|
|
"epoch": 0.07447393412305693,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.8103,
|
|
"mean_token_accuracy": 0.08842997774481773,
|
|
"num_tokens": 1459582.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 8.07224726676941,
|
|
"epoch": 0.0749737591842855,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.8624,
|
|
"mean_token_accuracy": 0.08552923649549485,
|
|
"num_tokens": 1468151.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 8.003483009338378,
|
|
"epoch": 0.07547358424551406,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.7729,
|
|
"mean_token_accuracy": 0.09006534814834595,
|
|
"num_tokens": 1479099.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 8.126268005371093,
|
|
"epoch": 0.07597340930674264,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.873,
|
|
"mean_token_accuracy": 0.09025009647011757,
|
|
"num_tokens": 1488060.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 8.105157661437989,
|
|
"epoch": 0.07647323436797121,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.8384,
|
|
"mean_token_accuracy": 0.08615945875644684,
|
|
"num_tokens": 1498230.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 8.004671478271485,
|
|
"epoch": 0.07697305942919978,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.8242,
|
|
"mean_token_accuracy": 0.08832557871937752,
|
|
"num_tokens": 1507962.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 8.232299852371217,
|
|
"epoch": 0.07747288449042836,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.9545,
|
|
"mean_token_accuracy": 0.08387064412236214,
|
|
"num_tokens": 1518607.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 8.105499935150146,
|
|
"epoch": 0.07797270955165692,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.7776,
|
|
"mean_token_accuracy": 0.0910347007215023,
|
|
"num_tokens": 1527166.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 8.119454288482666,
|
|
"epoch": 0.07847253461288549,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.8703,
|
|
"mean_token_accuracy": 0.08840905651450157,
|
|
"num_tokens": 1536683.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 8.03728346824646,
|
|
"epoch": 0.07897235967411406,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.7568,
|
|
"mean_token_accuracy": 0.09142104163765907,
|
|
"num_tokens": 1546243.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 8.057508659362792,
|
|
"epoch": 0.07947218473534264,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.7846,
|
|
"mean_token_accuracy": 0.0889986164867878,
|
|
"num_tokens": 1555916.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.945902585983276,
|
|
"epoch": 0.0799720097965712,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.6987,
|
|
"mean_token_accuracy": 0.094569893181324,
|
|
"num_tokens": 1566617.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.919804286956787,
|
|
"epoch": 0.08047183485779977,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.7405,
|
|
"mean_token_accuracy": 0.08826911225914955,
|
|
"num_tokens": 1575413.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 8.038426303863526,
|
|
"epoch": 0.08097165991902834,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.7465,
|
|
"mean_token_accuracy": 0.09363904744386672,
|
|
"num_tokens": 1586863.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 8.045948362350464,
|
|
"epoch": 0.08147148498025691,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.8292,
|
|
"mean_token_accuracy": 0.09168099835515023,
|
|
"num_tokens": 1597685.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 8.02879524230957,
|
|
"epoch": 0.08197131004148547,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.7286,
|
|
"mean_token_accuracy": 0.0919680006802082,
|
|
"num_tokens": 1607103.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 8.00586986541748,
|
|
"epoch": 0.08247113510271405,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.7797,
|
|
"mean_token_accuracy": 0.09101953282952309,
|
|
"num_tokens": 1617919.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 8.015330362319947,
|
|
"epoch": 0.08297096016394262,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.8094,
|
|
"mean_token_accuracy": 0.09074466563761234,
|
|
"num_tokens": 1627293.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 8.053974103927612,
|
|
"epoch": 0.08347078522517119,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.8196,
|
|
"mean_token_accuracy": 0.08691304028034211,
|
|
"num_tokens": 1638440.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.915004444122315,
|
|
"epoch": 0.08397061028639977,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.7225,
|
|
"mean_token_accuracy": 0.09402119815349579,
|
|
"num_tokens": 1648148.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.9461534976959225,
|
|
"epoch": 0.08447043534762833,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.75,
|
|
"mean_token_accuracy": 0.0946639321744442,
|
|
"num_tokens": 1658334.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 8.101148223876953,
|
|
"epoch": 0.0849702604088569,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.7948,
|
|
"mean_token_accuracy": 0.08722254931926728,
|
|
"num_tokens": 1667963.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.956143712997436,
|
|
"epoch": 0.08547008547008547,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.6999,
|
|
"mean_token_accuracy": 0.09139039814472198,
|
|
"num_tokens": 1677270.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.8967897415161135,
|
|
"epoch": 0.08596991053131404,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.6758,
|
|
"mean_token_accuracy": 0.09658752381801605,
|
|
"num_tokens": 1686770.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.969452381134033,
|
|
"epoch": 0.0864697355925426,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.7521,
|
|
"mean_token_accuracy": 0.09170125946402549,
|
|
"num_tokens": 1697507.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.872890663146973,
|
|
"epoch": 0.08696956065377118,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.6841,
|
|
"mean_token_accuracy": 0.09290253594517708,
|
|
"num_tokens": 1708253.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 8.000007724761963,
|
|
"epoch": 0.08746938571499975,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.7539,
|
|
"mean_token_accuracy": 0.08929965049028396,
|
|
"num_tokens": 1717827.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.832028770446778,
|
|
"epoch": 0.08796921077622832,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5521,
|
|
"mean_token_accuracy": 0.10026744902133941,
|
|
"num_tokens": 1726678.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.88534722328186,
|
|
"epoch": 0.0884690358374569,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.7009,
|
|
"mean_token_accuracy": 0.09250150695443153,
|
|
"num_tokens": 1737518.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.921040725708008,
|
|
"epoch": 0.08896886089868546,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5908,
|
|
"mean_token_accuracy": 0.10193264484405518,
|
|
"num_tokens": 1746338.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.840936279296875,
|
|
"epoch": 0.08946868595991403,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.6714,
|
|
"mean_token_accuracy": 0.09223530441522598,
|
|
"num_tokens": 1756538.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.921091079711914,
|
|
"epoch": 0.0899685110211426,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.6709,
|
|
"mean_token_accuracy": 0.09273412451148033,
|
|
"num_tokens": 1766722.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.847049188613892,
|
|
"epoch": 0.09046833608237118,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.6012,
|
|
"mean_token_accuracy": 0.09680447876453399,
|
|
"num_tokens": 1777039.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.876219177246094,
|
|
"epoch": 0.09096816114359974,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.7057,
|
|
"mean_token_accuracy": 0.09129747226834298,
|
|
"num_tokens": 1787621.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.934077644348145,
|
|
"epoch": 0.09146798620482831,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.7262,
|
|
"mean_token_accuracy": 0.09550910741090775,
|
|
"num_tokens": 1798345.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.863957357406616,
|
|
"epoch": 0.09196781126605688,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.6484,
|
|
"mean_token_accuracy": 0.09627768769860268,
|
|
"num_tokens": 1807994.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.814187479019165,
|
|
"epoch": 0.09246763632728545,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.6025,
|
|
"mean_token_accuracy": 0.09041696637868882,
|
|
"num_tokens": 1817965.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.898249673843384,
|
|
"epoch": 0.09296746138851401,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.651,
|
|
"mean_token_accuracy": 0.09574145302176476,
|
|
"num_tokens": 1827094.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.875781297683716,
|
|
"epoch": 0.09346728644974259,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.62,
|
|
"mean_token_accuracy": 0.09457285776734352,
|
|
"num_tokens": 1837754.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.698305368423462,
|
|
"epoch": 0.09396711151097116,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.607,
|
|
"mean_token_accuracy": 0.09555240049958229,
|
|
"num_tokens": 1847500.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.8827156066894535,
|
|
"epoch": 0.09446693657219973,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.6649,
|
|
"mean_token_accuracy": 0.08770858943462372,
|
|
"num_tokens": 1857663.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.890668249130249,
|
|
"epoch": 0.0949667616334283,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.7266,
|
|
"mean_token_accuracy": 0.09187851324677468,
|
|
"num_tokens": 1867752.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.822569274902344,
|
|
"epoch": 0.09546658669465687,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.6389,
|
|
"mean_token_accuracy": 0.09240206554532052,
|
|
"num_tokens": 1877479.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.813203954696656,
|
|
"epoch": 0.09596641175588544,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.6736,
|
|
"mean_token_accuracy": 0.09363591149449349,
|
|
"num_tokens": 1887944.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.859562635421753,
|
|
"epoch": 0.09646623681711401,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.5872,
|
|
"mean_token_accuracy": 0.09667022675275802,
|
|
"num_tokens": 1896991.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.771055555343628,
|
|
"epoch": 0.09696606187834259,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.5995,
|
|
"mean_token_accuracy": 0.10148105099797249,
|
|
"num_tokens": 1907216.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.911443471908569,
|
|
"epoch": 0.09746588693957114,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.6083,
|
|
"mean_token_accuracy": 0.09393106177449226,
|
|
"num_tokens": 1917206.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.743614768981933,
|
|
"epoch": 0.09796571200079972,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.6854,
|
|
"mean_token_accuracy": 0.09089522212743759,
|
|
"num_tokens": 1929669.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.8640203952789305,
|
|
"epoch": 0.09846553706202829,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.5627,
|
|
"mean_token_accuracy": 0.09610607698559762,
|
|
"num_tokens": 1939881.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.689713811874389,
|
|
"epoch": 0.09896536212325686,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5888,
|
|
"mean_token_accuracy": 0.09820158705115319,
|
|
"num_tokens": 1949423.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.818390989303589,
|
|
"epoch": 0.09946518718448542,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.6107,
|
|
"mean_token_accuracy": 0.09391192570328713,
|
|
"num_tokens": 1959214.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.867146348953247,
|
|
"epoch": 0.099965012245714,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.5943,
|
|
"mean_token_accuracy": 0.09512368217110634,
|
|
"num_tokens": 1968963.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.708599042892456,
|
|
"epoch": 0.10046483730694257,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999999981884985,
|
|
"loss": 7.5637,
|
|
"mean_token_accuracy": 0.0960096038877964,
|
|
"num_tokens": 1979551.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.774115419387817,
|
|
"epoch": 0.10096466236817114,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999999908292739,
|
|
"loss": 7.5299,
|
|
"mean_token_accuracy": 0.09881004914641381,
|
|
"num_tokens": 1988432.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.747634410858154,
|
|
"epoch": 0.10146448742939972,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999999778091072,
|
|
"loss": 7.5714,
|
|
"mean_token_accuracy": 0.09639670103788375,
|
|
"num_tokens": 1997509.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.858575534820557,
|
|
"epoch": 0.10196431249062828,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999999591279991,
|
|
"loss": 7.7288,
|
|
"mean_token_accuracy": 0.09117813184857368,
|
|
"num_tokens": 2007406.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.8197108745574955,
|
|
"epoch": 0.10246413755185685,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00049999993478595,
|
|
"loss": 7.5532,
|
|
"mean_token_accuracy": 0.09217148125171662,
|
|
"num_tokens": 2016760.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.714381504058838,
|
|
"epoch": 0.10296396261308542,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999999047829604,
|
|
"loss": 7.5862,
|
|
"mean_token_accuracy": 0.0973098672926426,
|
|
"num_tokens": 2027776.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.711545372009278,
|
|
"epoch": 0.103463787674314,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499999869119031,
|
|
"loss": 7.525,
|
|
"mean_token_accuracy": 0.09987044408917427,
|
|
"num_tokens": 2036955.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.708169794082641,
|
|
"epoch": 0.10396361273554255,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999998277941629,
|
|
"loss": 7.4481,
|
|
"mean_token_accuracy": 0.09874232336878777,
|
|
"num_tokens": 2046118.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.663513088226319,
|
|
"epoch": 0.10446343779677113,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999997808083569,
|
|
"loss": 7.5182,
|
|
"mean_token_accuracy": 0.09265113547444344,
|
|
"num_tokens": 2056266.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.766235733032227,
|
|
"epoch": 0.1049632628579997,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004999997281616144,
|
|
"loss": 7.5564,
|
|
"mean_token_accuracy": 0.09874644502997398,
|
|
"num_tokens": 2067416.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.742217063903809,
|
|
"epoch": 0.10546308791922827,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999996698539366,
|
|
"loss": 7.5323,
|
|
"mean_token_accuracy": 0.09508499130606651,
|
|
"num_tokens": 2076926.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.7295256614685055,
|
|
"epoch": 0.10596291298045685,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499999605885325,
|
|
"loss": 7.5473,
|
|
"mean_token_accuracy": 0.10004104599356652,
|
|
"num_tokens": 2088017.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.6699363708496096,
|
|
"epoch": 0.1064627380416854,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999995362557813,
|
|
"loss": 7.5634,
|
|
"mean_token_accuracy": 0.09938672706484794,
|
|
"num_tokens": 2097249.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.716280460357666,
|
|
"epoch": 0.10696256310291398,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499999460965307,
|
|
"loss": 7.4976,
|
|
"mean_token_accuracy": 0.09836835712194443,
|
|
"num_tokens": 2105778.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.753679323196411,
|
|
"epoch": 0.10746238816414255,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999993800139043,
|
|
"loss": 7.5311,
|
|
"mean_token_accuracy": 0.09935536906123162,
|
|
"num_tokens": 2114710.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.64402494430542,
|
|
"epoch": 0.10796221322537113,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999992934015752,
|
|
"loss": 7.4801,
|
|
"mean_token_accuracy": 0.09851668924093246,
|
|
"num_tokens": 2124002.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.777598571777344,
|
|
"epoch": 0.10846203828659969,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999992011283215,
|
|
"loss": 7.5611,
|
|
"mean_token_accuracy": 0.09994415789842606,
|
|
"num_tokens": 2134592.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.567396640777588,
|
|
"epoch": 0.10896186334782826,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999991031941459,
|
|
"loss": 7.4156,
|
|
"mean_token_accuracy": 0.10332436636090278,
|
|
"num_tokens": 2143641.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.74939751625061,
|
|
"epoch": 0.10946168840905683,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999989995990509,
|
|
"loss": 7.5304,
|
|
"mean_token_accuracy": 0.09657632037997246,
|
|
"num_tokens": 2154738.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.590510129928589,
|
|
"epoch": 0.1099615134702854,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999988903430389,
|
|
"loss": 7.4692,
|
|
"mean_token_accuracy": 0.10435619726777076,
|
|
"num_tokens": 2164365.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.6218901634216305,
|
|
"epoch": 0.11046133853151396,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999987754261126,
|
|
"loss": 7.3518,
|
|
"mean_token_accuracy": 0.1033005841076374,
|
|
"num_tokens": 2173185.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.6353357315063475,
|
|
"epoch": 0.11096116359274254,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999986548482751,
|
|
"loss": 7.4808,
|
|
"mean_token_accuracy": 0.1050780363380909,
|
|
"num_tokens": 2184717.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.639813375473023,
|
|
"epoch": 0.11146098865397111,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999985286095293,
|
|
"loss": 7.4457,
|
|
"mean_token_accuracy": 0.10765890702605248,
|
|
"num_tokens": 2193681.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.63369607925415,
|
|
"epoch": 0.11196081371519968,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999983967098785,
|
|
"loss": 7.4596,
|
|
"mean_token_accuracy": 0.09970392286777496,
|
|
"num_tokens": 2203366.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.701994895935059,
|
|
"epoch": 0.11246063877642826,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999982591493258,
|
|
"loss": 7.459,
|
|
"mean_token_accuracy": 0.10135932490229607,
|
|
"num_tokens": 2213982.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.644823122024536,
|
|
"epoch": 0.11296046383765682,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999981159278749,
|
|
"loss": 7.4974,
|
|
"mean_token_accuracy": 0.09790250882506371,
|
|
"num_tokens": 2222897.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.627260303497314,
|
|
"epoch": 0.11346028889888539,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999979670455292,
|
|
"loss": 7.446,
|
|
"mean_token_accuracy": 0.10249497219920159,
|
|
"num_tokens": 2232472.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.57940993309021,
|
|
"epoch": 0.11396011396011396,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004999978125022926,
|
|
"loss": 7.489,
|
|
"mean_token_accuracy": 0.10549622848629951,
|
|
"num_tokens": 2242666.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.741400814056396,
|
|
"epoch": 0.11445993902134254,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499997652298169,
|
|
"loss": 7.568,
|
|
"mean_token_accuracy": 0.09314922839403153,
|
|
"num_tokens": 2254586.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.5815185546875,
|
|
"epoch": 0.1149597640825711,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999974864331624,
|
|
"loss": 7.3888,
|
|
"mean_token_accuracy": 0.10329079777002334,
|
|
"num_tokens": 2264619.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.781215858459473,
|
|
"epoch": 0.11545958914379967,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999973149072768,
|
|
"loss": 7.4773,
|
|
"mean_token_accuracy": 0.0925179947167635,
|
|
"num_tokens": 2274114.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.433381843566894,
|
|
"epoch": 0.11595941420502824,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999971377205167,
|
|
"loss": 7.3847,
|
|
"mean_token_accuracy": 0.10573233887553216,
|
|
"num_tokens": 2283887.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.647586059570313,
|
|
"epoch": 0.11645923926625681,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999969548728864,
|
|
"loss": 7.3971,
|
|
"mean_token_accuracy": 0.1060848891735077,
|
|
"num_tokens": 2293090.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.600371837615967,
|
|
"epoch": 0.11695906432748537,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999967663643908,
|
|
"loss": 7.5024,
|
|
"mean_token_accuracy": 0.09858990088105202,
|
|
"num_tokens": 2302828.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.6886982917785645,
|
|
"epoch": 0.11745888938871395,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999965721950345,
|
|
"loss": 7.3914,
|
|
"mean_token_accuracy": 0.10518279448151588,
|
|
"num_tokens": 2311415.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.562065887451172,
|
|
"epoch": 0.11795871444994252,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999963723648222,
|
|
"loss": 7.4684,
|
|
"mean_token_accuracy": 0.09992958679795265,
|
|
"num_tokens": 2321512.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.6449196338653564,
|
|
"epoch": 0.1184585395111711,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000499996166873759,
|
|
"loss": 7.3802,
|
|
"mean_token_accuracy": 0.10009320005774498,
|
|
"num_tokens": 2331398.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.582540845870971,
|
|
"epoch": 0.11895836457239967,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999959557218502,
|
|
"loss": 7.378,
|
|
"mean_token_accuracy": 0.10733285993337631,
|
|
"num_tokens": 2341751.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.647002792358398,
|
|
"epoch": 0.11945818963362823,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999957389091012,
|
|
"loss": 7.3963,
|
|
"mean_token_accuracy": 0.09711159616708756,
|
|
"num_tokens": 2352538.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.58666090965271,
|
|
"epoch": 0.1199580146948568,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999955164355171,
|
|
"loss": 7.4581,
|
|
"mean_token_accuracy": 0.09764329642057419,
|
|
"num_tokens": 2362433.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.584440898895264,
|
|
"epoch": 0.12045783975608537,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999952883011038,
|
|
"loss": 7.4382,
|
|
"mean_token_accuracy": 0.09720063619315625,
|
|
"num_tokens": 2372431.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.566992712020874,
|
|
"epoch": 0.12095766481731395,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499995054505867,
|
|
"loss": 7.4337,
|
|
"mean_token_accuracy": 0.10857149809598923,
|
|
"num_tokens": 2382483.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.723560905456543,
|
|
"epoch": 0.1214574898785425,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999948150498124,
|
|
"loss": 7.4276,
|
|
"mean_token_accuracy": 0.09219059199094773,
|
|
"num_tokens": 2392835.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.536263561248779,
|
|
"epoch": 0.12195731493977108,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999945699329462,
|
|
"loss": 7.3764,
|
|
"mean_token_accuracy": 0.09699192494153977,
|
|
"num_tokens": 2402305.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.461746692657471,
|
|
"epoch": 0.12245714000099965,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999943191552745,
|
|
"loss": 7.3208,
|
|
"mean_token_accuracy": 0.10833752900362015,
|
|
"num_tokens": 2412093.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.578193950653076,
|
|
"epoch": 0.12295696506222822,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999940627168037,
|
|
"loss": 7.2861,
|
|
"mean_token_accuracy": 0.10602069944143296,
|
|
"num_tokens": 2421349.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.4720502376556395,
|
|
"epoch": 0.12345679012345678,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00049999380061754,
|
|
"loss": 7.2871,
|
|
"mean_token_accuracy": 0.11482211500406266,
|
|
"num_tokens": 2430672.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.4783854484558105,
|
|
"epoch": 0.12395661518468536,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999935328574903,
|
|
"loss": 7.3577,
|
|
"mean_token_accuracy": 0.10103069022297859,
|
|
"num_tokens": 2439355.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.5710039138793945,
|
|
"epoch": 0.12445644024591393,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999932594366613,
|
|
"loss": 7.3519,
|
|
"mean_token_accuracy": 0.10859728455543519,
|
|
"num_tokens": 2448860.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.42594199180603,
|
|
"epoch": 0.1249562653071425,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999929803550597,
|
|
"loss": 7.3812,
|
|
"mean_token_accuracy": 0.10127258449792861,
|
|
"num_tokens": 2458276.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.5558329105377195,
|
|
"epoch": 0.12545609036837108,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999926956126927,
|
|
"loss": 7.2995,
|
|
"mean_token_accuracy": 0.10753061398863792,
|
|
"num_tokens": 2467791.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.481610918045044,
|
|
"epoch": 0.12595591542959964,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999924052095672,
|
|
"loss": 7.4297,
|
|
"mean_token_accuracy": 0.10026444271206855,
|
|
"num_tokens": 2477549.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.551158761978149,
|
|
"epoch": 0.12645574049082822,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999921091456907,
|
|
"loss": 7.3405,
|
|
"mean_token_accuracy": 0.09958168342709542,
|
|
"num_tokens": 2487549.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.575411987304688,
|
|
"epoch": 0.12695556555205678,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999918074210707,
|
|
"loss": 7.3359,
|
|
"mean_token_accuracy": 0.10639088824391366,
|
|
"num_tokens": 2496799.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.616192436218261,
|
|
"epoch": 0.12745539061328534,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999915000357147,
|
|
"loss": 7.4325,
|
|
"mean_token_accuracy": 0.10289375483989716,
|
|
"num_tokens": 2505795.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.4822615623474125,
|
|
"epoch": 0.12795521567451393,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999911869896305,
|
|
"loss": 7.2474,
|
|
"mean_token_accuracy": 0.10294352620840072,
|
|
"num_tokens": 2515289.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.597700119018555,
|
|
"epoch": 0.1284550407357425,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999908682828258,
|
|
"loss": 7.4296,
|
|
"mean_token_accuracy": 0.09725573807954788,
|
|
"num_tokens": 2524563.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.434320259094238,
|
|
"epoch": 0.12895486579697105,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999905439153089,
|
|
"loss": 7.3241,
|
|
"mean_token_accuracy": 0.10576057732105255,
|
|
"num_tokens": 2534057.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.505590295791626,
|
|
"epoch": 0.12945469085819963,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999902138870878,
|
|
"loss": 7.3078,
|
|
"mean_token_accuracy": 0.10811854675412177,
|
|
"num_tokens": 2544646.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.574274253845215,
|
|
"epoch": 0.1299545159194282,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999898781981707,
|
|
"loss": 7.3776,
|
|
"mean_token_accuracy": 0.10291878581047058,
|
|
"num_tokens": 2553843.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.4811914443969725,
|
|
"epoch": 0.13045434098065678,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999895368485662,
|
|
"loss": 7.4132,
|
|
"mean_token_accuracy": 0.09877474531531334,
|
|
"num_tokens": 2564061.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.492168235778808,
|
|
"epoch": 0.13095416604188534,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000499989189838283,
|
|
"loss": 7.2787,
|
|
"mean_token_accuracy": 0.10895716845989227,
|
|
"num_tokens": 2574758.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.485337257385254,
|
|
"epoch": 0.1314539911031139,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999888371673295,
|
|
"loss": 7.35,
|
|
"mean_token_accuracy": 0.10393542721867562,
|
|
"num_tokens": 2585213.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.545273399353027,
|
|
"epoch": 0.13195381616434249,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004999884788357147,
|
|
"loss": 7.3947,
|
|
"mean_token_accuracy": 0.10360251069068908,
|
|
"num_tokens": 2595511.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.474040794372558,
|
|
"epoch": 0.13245364122557104,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999881148434478,
|
|
"loss": 7.3246,
|
|
"mean_token_accuracy": 0.1028148852288723,
|
|
"num_tokens": 2605930.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.545388984680176,
|
|
"epoch": 0.13295346628679963,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999877451905378,
|
|
"loss": 7.2089,
|
|
"mean_token_accuracy": 0.11397390365600586,
|
|
"num_tokens": 2614224.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.5077166080474855,
|
|
"epoch": 0.1334532913480282,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999873698769941,
|
|
"loss": 7.421,
|
|
"mean_token_accuracy": 0.09980487525463104,
|
|
"num_tokens": 2624879.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.493503904342651,
|
|
"epoch": 0.13395311640925675,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000499986988902826,
|
|
"loss": 7.2739,
|
|
"mean_token_accuracy": 0.10615406557917595,
|
|
"num_tokens": 2633836.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.377463626861572,
|
|
"epoch": 0.13445294147048534,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999866022680431,
|
|
"loss": 7.3316,
|
|
"mean_token_accuracy": 0.10316964089870453,
|
|
"num_tokens": 2643754.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.616025114059449,
|
|
"epoch": 0.1349527665317139,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999862099726552,
|
|
"loss": 7.3922,
|
|
"mean_token_accuracy": 0.09743982404470444,
|
|
"num_tokens": 2652708.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.4626898765563965,
|
|
"epoch": 0.13545259159294248,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999858120166721,
|
|
"loss": 7.352,
|
|
"mean_token_accuracy": 0.1009536899626255,
|
|
"num_tokens": 2662353.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.507920217514038,
|
|
"epoch": 0.13595241665417104,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999854084001039,
|
|
"loss": 7.2551,
|
|
"mean_token_accuracy": 0.10801847130060196,
|
|
"num_tokens": 2672376.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.4212922096252445,
|
|
"epoch": 0.1364522417153996,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999849991229608,
|
|
"loss": 7.3581,
|
|
"mean_token_accuracy": 0.10042979195713997,
|
|
"num_tokens": 2682115.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.520157718658448,
|
|
"epoch": 0.1369520667766282,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499984584185253,
|
|
"loss": 7.3351,
|
|
"mean_token_accuracy": 0.10273827612400055,
|
|
"num_tokens": 2692429.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.495982551574707,
|
|
"epoch": 0.13745189183785675,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999841635869909,
|
|
"loss": 7.2832,
|
|
"mean_token_accuracy": 0.10646675527095795,
|
|
"num_tokens": 2701373.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.59482774734497,
|
|
"epoch": 0.1379517168990853,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999837373281852,
|
|
"loss": 7.4542,
|
|
"mean_token_accuracy": 0.10424424782395363,
|
|
"num_tokens": 2710717.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.465184688568115,
|
|
"epoch": 0.1384515419603139,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999833054088465,
|
|
"loss": 7.315,
|
|
"mean_token_accuracy": 0.10450968667864799,
|
|
"num_tokens": 2720936.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.52575716972351,
|
|
"epoch": 0.13895136702154245,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999828678289858,
|
|
"loss": 7.2724,
|
|
"mean_token_accuracy": 0.11038048714399337,
|
|
"num_tokens": 2730504.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.384076833724976,
|
|
"epoch": 0.13945119208277104,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499982424588614,
|
|
"loss": 7.2831,
|
|
"mean_token_accuracy": 0.1060723252594471,
|
|
"num_tokens": 2739069.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.52955813407898,
|
|
"epoch": 0.1399510171439996,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999819756877422,
|
|
"loss": 7.3542,
|
|
"mean_token_accuracy": 0.09965510442852973,
|
|
"num_tokens": 2748941.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.469629859924316,
|
|
"epoch": 0.14045084220522816,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999815211263819,
|
|
"loss": 7.3046,
|
|
"mean_token_accuracy": 0.10589431822299958,
|
|
"num_tokens": 2758915.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.496847534179688,
|
|
"epoch": 0.14095066726645675,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999810609045444,
|
|
"loss": 7.3413,
|
|
"mean_token_accuracy": 0.10205910354852676,
|
|
"num_tokens": 2768972.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.4970423698425295,
|
|
"epoch": 0.1414504923276853,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999805950222412,
|
|
"loss": 7.1917,
|
|
"mean_token_accuracy": 0.10759721770882606,
|
|
"num_tokens": 2777989.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.3449572086334225,
|
|
"epoch": 0.1419503173889139,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999801234794843,
|
|
"loss": 7.2963,
|
|
"mean_token_accuracy": 0.10734140872955322,
|
|
"num_tokens": 2788227.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.547876358032227,
|
|
"epoch": 0.14245014245014245,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999796462762853,
|
|
"loss": 7.2899,
|
|
"mean_token_accuracy": 0.10559340417385102,
|
|
"num_tokens": 2797729.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.433524799346924,
|
|
"epoch": 0.142949967511371,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999791634126562,
|
|
"loss": 7.2836,
|
|
"mean_token_accuracy": 0.10649279952049255,
|
|
"num_tokens": 2807215.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.472461366653443,
|
|
"epoch": 0.1434497925725996,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999786748886094,
|
|
"loss": 7.202,
|
|
"mean_token_accuracy": 0.10329415574669838,
|
|
"num_tokens": 2816945.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.3642683029174805,
|
|
"epoch": 0.14394961763382816,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999781807041569,
|
|
"loss": 7.2831,
|
|
"mean_token_accuracy": 0.10314463749527931,
|
|
"num_tokens": 2827072.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.438300848007202,
|
|
"epoch": 0.14444944269505672,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999776808593113,
|
|
"loss": 7.2172,
|
|
"mean_token_accuracy": 0.1053712360560894,
|
|
"num_tokens": 2837928.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.362199783325195,
|
|
"epoch": 0.1449492677562853,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999771753540852,
|
|
"loss": 7.2508,
|
|
"mean_token_accuracy": 0.10845730304718018,
|
|
"num_tokens": 2847861.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.463563776016235,
|
|
"epoch": 0.14544909281751386,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999766641884912,
|
|
"loss": 7.248,
|
|
"mean_token_accuracy": 0.10768592208623887,
|
|
"num_tokens": 2857885.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.49535756111145,
|
|
"epoch": 0.14594891787874245,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999761473625422,
|
|
"loss": 7.2817,
|
|
"mean_token_accuracy": 0.1021933138370514,
|
|
"num_tokens": 2867004.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.361451148986816,
|
|
"epoch": 0.146448742939971,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999756248762513,
|
|
"loss": 7.1422,
|
|
"mean_token_accuracy": 0.10897669419646264,
|
|
"num_tokens": 2876285.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.437837982177735,
|
|
"epoch": 0.14694856800119957,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999750967296315,
|
|
"loss": 7.2794,
|
|
"mean_token_accuracy": 0.10400166437029838,
|
|
"num_tokens": 2886967.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.400296020507812,
|
|
"epoch": 0.14744839306242816,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999745629226962,
|
|
"loss": 7.2407,
|
|
"mean_token_accuracy": 0.10448872894048691,
|
|
"num_tokens": 2897137.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.402553701400757,
|
|
"epoch": 0.14794821812365672,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999740234554589,
|
|
"loss": 7.2226,
|
|
"mean_token_accuracy": 0.11351158022880554,
|
|
"num_tokens": 2907658.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.393366098403931,
|
|
"epoch": 0.1484480431848853,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999734783279327,
|
|
"loss": 7.2207,
|
|
"mean_token_accuracy": 0.10734575539827347,
|
|
"num_tokens": 2917801.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.441537857055664,
|
|
"epoch": 0.14894786824611386,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499972927540132,
|
|
"loss": 7.2977,
|
|
"mean_token_accuracy": 0.10554568693041802,
|
|
"num_tokens": 2927188.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.3817533493042,
|
|
"epoch": 0.14944769330734242,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999723710920703,
|
|
"loss": 7.1798,
|
|
"mean_token_accuracy": 0.10319826304912567,
|
|
"num_tokens": 2936781.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.3468931198120115,
|
|
"epoch": 0.149947518368571,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999718089837616,
|
|
"loss": 7.1681,
|
|
"mean_token_accuracy": 0.10979624688625336,
|
|
"num_tokens": 2946426.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.339980173110962,
|
|
"epoch": 0.15044734342979957,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00049997124121522,
|
|
"loss": 7.2007,
|
|
"mean_token_accuracy": 0.10412992984056473,
|
|
"num_tokens": 2956246.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.4069291114807125,
|
|
"epoch": 0.15094716849102813,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999706677864599,
|
|
"loss": 7.2068,
|
|
"mean_token_accuracy": 0.1070282831788063,
|
|
"num_tokens": 2965930.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.319704008102417,
|
|
"epoch": 0.15144699355225671,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999700886974958,
|
|
"loss": 7.1387,
|
|
"mean_token_accuracy": 0.11144708767533303,
|
|
"num_tokens": 2975975.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.368978881835938,
|
|
"epoch": 0.15194681861348527,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000499969503948342,
|
|
"loss": 7.1486,
|
|
"mean_token_accuracy": 0.10778968706727028,
|
|
"num_tokens": 2985343.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.321302700042724,
|
|
"epoch": 0.15244664367471386,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999689135390134,
|
|
"loss": 7.1615,
|
|
"mean_token_accuracy": 0.11044318154454232,
|
|
"num_tokens": 2994420.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.383437299728394,
|
|
"epoch": 0.15294646873594242,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999683174695249,
|
|
"loss": 7.2869,
|
|
"mean_token_accuracy": 0.10852601900696754,
|
|
"num_tokens": 3004447.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.49035234451294,
|
|
"epoch": 0.15344629379717098,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999677157398914,
|
|
"loss": 7.177,
|
|
"mean_token_accuracy": 0.10626579597592353,
|
|
"num_tokens": 3014596.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.332225370407104,
|
|
"epoch": 0.15394611885839957,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999671083501281,
|
|
"loss": 7.111,
|
|
"mean_token_accuracy": 0.1120404727756977,
|
|
"num_tokens": 3024528.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.32868595123291,
|
|
"epoch": 0.15444594391962813,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999664953002502,
|
|
"loss": 7.2234,
|
|
"mean_token_accuracy": 0.10404160246253014,
|
|
"num_tokens": 3035233.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.376626300811767,
|
|
"epoch": 0.1549457689808567,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499965876590273,
|
|
"loss": 7.2365,
|
|
"mean_token_accuracy": 0.10801902562379836,
|
|
"num_tokens": 3045890.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.326313638687134,
|
|
"epoch": 0.15544559404208527,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999652522202125,
|
|
"loss": 7.1306,
|
|
"mean_token_accuracy": 0.11243928968906403,
|
|
"num_tokens": 3056640.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.366794013977051,
|
|
"epoch": 0.15594541910331383,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999646221900839,
|
|
"loss": 7.1631,
|
|
"mean_token_accuracy": 0.11329249888658524,
|
|
"num_tokens": 3066938.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.35032844543457,
|
|
"epoch": 0.15644524416454242,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004999639864999034,
|
|
"loss": 7.236,
|
|
"mean_token_accuracy": 0.11088693737983704,
|
|
"num_tokens": 3076791.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.395107746124268,
|
|
"epoch": 0.15694506922577098,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499963345149687,
|
|
"loss": 7.1203,
|
|
"mean_token_accuracy": 0.10925953909754753,
|
|
"num_tokens": 3087532.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.281245279312134,
|
|
"epoch": 0.15744489428699954,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999626981394506,
|
|
"loss": 7.2466,
|
|
"mean_token_accuracy": 0.10709367841482162,
|
|
"num_tokens": 3097465.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.441256141662597,
|
|
"epoch": 0.15794471934822812,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999620454692106,
|
|
"loss": 7.1655,
|
|
"mean_token_accuracy": 0.10919474661350251,
|
|
"num_tokens": 3107162.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.253028106689453,
|
|
"epoch": 0.15844454440945668,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999613871389836,
|
|
"loss": 7.1456,
|
|
"mean_token_accuracy": 0.11203786730766296,
|
|
"num_tokens": 3117154.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.32353892326355,
|
|
"epoch": 0.15894436947068527,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999607231487858,
|
|
"loss": 7.1594,
|
|
"mean_token_accuracy": 0.10915421843528747,
|
|
"num_tokens": 3127635.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.413996791839599,
|
|
"epoch": 0.15944419453191383,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000499960053498634,
|
|
"loss": 7.1971,
|
|
"mean_token_accuracy": 0.10433795824646949,
|
|
"num_tokens": 3137806.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.3293005466461185,
|
|
"epoch": 0.1599440195931424,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999593781885454,
|
|
"loss": 7.2291,
|
|
"mean_token_accuracy": 0.10129423663020135,
|
|
"num_tokens": 3147704.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.344339752197266,
|
|
"epoch": 0.16044384465437098,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999586972185366,
|
|
"loss": 7.177,
|
|
"mean_token_accuracy": 0.10396260619163514,
|
|
"num_tokens": 3158486.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.358778429031372,
|
|
"epoch": 0.16094366971559954,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999580105886248,
|
|
"loss": 7.1816,
|
|
"mean_token_accuracy": 0.11292050555348396,
|
|
"num_tokens": 3168518.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.306634759902954,
|
|
"epoch": 0.16144349477682812,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999573182988275,
|
|
"loss": 7.093,
|
|
"mean_token_accuracy": 0.11131592839956284,
|
|
"num_tokens": 3177502.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.302649450302124,
|
|
"epoch": 0.16194331983805668,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999566203491619,
|
|
"loss": 7.132,
|
|
"mean_token_accuracy": 0.11160676255822181,
|
|
"num_tokens": 3187408.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.337405490875244,
|
|
"epoch": 0.16244314489928524,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999559167396455,
|
|
"loss": 7.182,
|
|
"mean_token_accuracy": 0.11259776130318641,
|
|
"num_tokens": 3196812.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.335812187194824,
|
|
"epoch": 0.16294296996051383,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999552074702963,
|
|
"loss": 7.0628,
|
|
"mean_token_accuracy": 0.11406400948762893,
|
|
"num_tokens": 3206159.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.254757928848266,
|
|
"epoch": 0.1634427950217424,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999544925411318,
|
|
"loss": 7.0451,
|
|
"mean_token_accuracy": 0.11323473453521729,
|
|
"num_tokens": 3217329.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.261689615249634,
|
|
"epoch": 0.16394262008297095,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999537719521703,
|
|
"loss": 7.1639,
|
|
"mean_token_accuracy": 0.11070609316229821,
|
|
"num_tokens": 3227558.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.421111106872559,
|
|
"epoch": 0.16444244514419953,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999530457034297,
|
|
"loss": 7.2582,
|
|
"mean_token_accuracy": 0.10368104577064514,
|
|
"num_tokens": 3237719.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.341942644119262,
|
|
"epoch": 0.1649422702054281,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999523137949283,
|
|
"loss": 7.1627,
|
|
"mean_token_accuracy": 0.11192248985171319,
|
|
"num_tokens": 3246483.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.233231067657471,
|
|
"epoch": 0.16544209526665668,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004999515762266846,
|
|
"loss": 7.0634,
|
|
"mean_token_accuracy": 0.11377293840050698,
|
|
"num_tokens": 3256955.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.347592067718506,
|
|
"epoch": 0.16594192032788524,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999508329987173,
|
|
"loss": 7.1488,
|
|
"mean_token_accuracy": 0.10765791982412339,
|
|
"num_tokens": 3266799.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.334767866134643,
|
|
"epoch": 0.1664417453891138,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999500841110447,
|
|
"loss": 7.1478,
|
|
"mean_token_accuracy": 0.10780449137091637,
|
|
"num_tokens": 3276595.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.245278692245483,
|
|
"epoch": 0.16694157045034239,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499949329563686,
|
|
"loss": 7.0769,
|
|
"mean_token_accuracy": 0.10730566009879113,
|
|
"num_tokens": 3286047.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.316799879074097,
|
|
"epoch": 0.16744139551157095,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00049994856935666,
|
|
"loss": 7.1646,
|
|
"mean_token_accuracy": 0.10785932466387749,
|
|
"num_tokens": 3295364.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.33067831993103,
|
|
"epoch": 0.16794122057279953,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499947803489986,
|
|
"loss": 7.0745,
|
|
"mean_token_accuracy": 0.11385716944932937,
|
|
"num_tokens": 3305319.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.214070272445679,
|
|
"epoch": 0.1684410456340281,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999470319636829,
|
|
"loss": 7.0871,
|
|
"mean_token_accuracy": 0.11497763246297836,
|
|
"num_tokens": 3315349.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.287170171737671,
|
|
"epoch": 0.16894087069525665,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999462547777705,
|
|
"loss": 7.1709,
|
|
"mean_token_accuracy": 0.10371953621506691,
|
|
"num_tokens": 3325603.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.254241466522217,
|
|
"epoch": 0.16944069575648524,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999454719322682,
|
|
"loss": 7.0624,
|
|
"mean_token_accuracy": 0.11126351356506348,
|
|
"num_tokens": 3334693.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.293223333358765,
|
|
"epoch": 0.1699405208177138,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999446834271958,
|
|
"loss": 7.0913,
|
|
"mean_token_accuracy": 0.10409749746322632,
|
|
"num_tokens": 3345259.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.280128622055054,
|
|
"epoch": 0.17044034587894236,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999438892625728,
|
|
"loss": 7.0757,
|
|
"mean_token_accuracy": 0.11194226369261742,
|
|
"num_tokens": 3354072.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.393798208236694,
|
|
"epoch": 0.17094017094017094,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999430894384198,
|
|
"loss": 7.1109,
|
|
"mean_token_accuracy": 0.1075687400996685,
|
|
"num_tokens": 3364881.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.227674198150635,
|
|
"epoch": 0.1714399960013995,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999422839547564,
|
|
"loss": 7.0544,
|
|
"mean_token_accuracy": 0.11126514673233032,
|
|
"num_tokens": 3374030.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.292581796646118,
|
|
"epoch": 0.1719398210626281,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999414728116029,
|
|
"loss": 7.0712,
|
|
"mean_token_accuracy": 0.10977467969059944,
|
|
"num_tokens": 3384009.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.3453587055206295,
|
|
"epoch": 0.17243964612385665,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999406560089798,
|
|
"loss": 7.1764,
|
|
"mean_token_accuracy": 0.10677952840924262,
|
|
"num_tokens": 3393660.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.311243486404419,
|
|
"epoch": 0.1729394711850852,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999398335469078,
|
|
"loss": 7.2011,
|
|
"mean_token_accuracy": 0.11048303991556167,
|
|
"num_tokens": 3404364.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.421372318267823,
|
|
"epoch": 0.1734392962463138,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999390054254074,
|
|
"loss": 7.0974,
|
|
"mean_token_accuracy": 0.10910709127783776,
|
|
"num_tokens": 3412990.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.14683895111084,
|
|
"epoch": 0.17393912130754235,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999381716444996,
|
|
"loss": 7.0533,
|
|
"mean_token_accuracy": 0.1123741738498211,
|
|
"num_tokens": 3422339.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.241878366470337,
|
|
"epoch": 0.17443894636877094,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999373322042052,
|
|
"loss": 7.0462,
|
|
"mean_token_accuracy": 0.11458217501640319,
|
|
"num_tokens": 3433093.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.317088603973389,
|
|
"epoch": 0.1749387714299995,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999364871045455,
|
|
"loss": 7.1223,
|
|
"mean_token_accuracy": 0.11018144264817238,
|
|
"num_tokens": 3443464.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.248693656921387,
|
|
"epoch": 0.17543859649122806,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004999356363455415,
|
|
"loss": 7.079,
|
|
"mean_token_accuracy": 0.10889410227537155,
|
|
"num_tokens": 3453380.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.283649063110351,
|
|
"epoch": 0.17593842155245665,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499934779927215,
|
|
"loss": 7.1213,
|
|
"mean_token_accuracy": 0.10985768139362335,
|
|
"num_tokens": 3463583.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.192403268814087,
|
|
"epoch": 0.1764382466136852,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999339178495872,
|
|
"loss": 7.0285,
|
|
"mean_token_accuracy": 0.11696212887763976,
|
|
"num_tokens": 3473640.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.219367408752442,
|
|
"epoch": 0.1769380716749138,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999330501126798,
|
|
"loss": 7.0867,
|
|
"mean_token_accuracy": 0.10979457870125771,
|
|
"num_tokens": 3483827.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.295134878158569,
|
|
"epoch": 0.17743789673614235,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999321767165148,
|
|
"loss": 7.0601,
|
|
"mean_token_accuracy": 0.11907404363155365,
|
|
"num_tokens": 3493763.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.286717557907105,
|
|
"epoch": 0.1779377217973709,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999312976611142,
|
|
"loss": 7.1599,
|
|
"mean_token_accuracy": 0.10770777985453606,
|
|
"num_tokens": 3503119.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.266967391967773,
|
|
"epoch": 0.1784375468585995,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999304129465,
|
|
"loss": 7.0697,
|
|
"mean_token_accuracy": 0.11251412183046341,
|
|
"num_tokens": 3513235.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.309150743484497,
|
|
"epoch": 0.17893737191982806,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999295225726945,
|
|
"loss": 7.1909,
|
|
"mean_token_accuracy": 0.10825281664729118,
|
|
"num_tokens": 3523633.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.242037582397461,
|
|
"epoch": 0.17943719698105662,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00049992862653972,
|
|
"loss": 7.0344,
|
|
"mean_token_accuracy": 0.10948573946952819,
|
|
"num_tokens": 3532888.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.249222183227539,
|
|
"epoch": 0.1799370220422852,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999277248475992,
|
|
"loss": 7.0524,
|
|
"mean_token_accuracy": 0.11693340837955475,
|
|
"num_tokens": 3541942.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.217988681793213,
|
|
"epoch": 0.18043684710351376,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999268174963547,
|
|
"loss": 7.0303,
|
|
"mean_token_accuracy": 0.11028838530182838,
|
|
"num_tokens": 3551544.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.225018787384033,
|
|
"epoch": 0.18093667216474235,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999259044860093,
|
|
"loss": 7.0354,
|
|
"mean_token_accuracy": 0.11371938288211822,
|
|
"num_tokens": 3561024.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.231437492370605,
|
|
"epoch": 0.1814364972259709,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999249858165861,
|
|
"loss": 7.0097,
|
|
"mean_token_accuracy": 0.11793867349624634,
|
|
"num_tokens": 3570962.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.105706644058228,
|
|
"epoch": 0.18193632228719947,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999240614881082,
|
|
"loss": 7.013,
|
|
"mean_token_accuracy": 0.1166162334382534,
|
|
"num_tokens": 3580669.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.306497097015381,
|
|
"epoch": 0.18243614734842806,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999231315005987,
|
|
"loss": 7.0338,
|
|
"mean_token_accuracy": 0.11162179484963416,
|
|
"num_tokens": 3590627.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.187291955947876,
|
|
"epoch": 0.18293597240965662,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999221958540811,
|
|
"loss": 7.0953,
|
|
"mean_token_accuracy": 0.10703551173210143,
|
|
"num_tokens": 3599872.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.294304513931275,
|
|
"epoch": 0.1834357974708852,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999212545485789,
|
|
"loss": 7.0767,
|
|
"mean_token_accuracy": 0.11386150121688843,
|
|
"num_tokens": 3609809.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.298921871185303,
|
|
"epoch": 0.18393562253211376,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999203075841159,
|
|
"loss": 7.0807,
|
|
"mean_token_accuracy": 0.11436494663357735,
|
|
"num_tokens": 3619385.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.1236320495605465,
|
|
"epoch": 0.18443544759334232,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999193549607157,
|
|
"loss": 6.9365,
|
|
"mean_token_accuracy": 0.1221968486905098,
|
|
"num_tokens": 3628867.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.1869734764099125,
|
|
"epoch": 0.1849352726545709,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999183966784026,
|
|
"loss": 6.957,
|
|
"mean_token_accuracy": 0.12176943942904472,
|
|
"num_tokens": 3637528.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.262745809555054,
|
|
"epoch": 0.18543509771579947,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004999174327372004,
|
|
"loss": 7.0832,
|
|
"mean_token_accuracy": 0.11131772547960281,
|
|
"num_tokens": 3648011.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.221261215209961,
|
|
"epoch": 0.18593492277702803,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999164631371335,
|
|
"loss": 7.0548,
|
|
"mean_token_accuracy": 0.11538778394460678,
|
|
"num_tokens": 3657648.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.252969884872437,
|
|
"epoch": 0.18643474783825661,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999154878782262,
|
|
"loss": 7.0716,
|
|
"mean_token_accuracy": 0.1120391070842743,
|
|
"num_tokens": 3666965.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.268553447723389,
|
|
"epoch": 0.18693457289948517,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000499914506960503,
|
|
"loss": 7.0761,
|
|
"mean_token_accuracy": 0.10713672935962677,
|
|
"num_tokens": 3676609.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.212067604064941,
|
|
"epoch": 0.18743439796071376,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999135203839889,
|
|
"loss": 6.958,
|
|
"mean_token_accuracy": 0.11653830632567405,
|
|
"num_tokens": 3685884.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.169482707977295,
|
|
"epoch": 0.18793422302194232,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999125281487084,
|
|
"loss": 7.0926,
|
|
"mean_token_accuracy": 0.11437800228595733,
|
|
"num_tokens": 3695436.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.212578773498535,
|
|
"epoch": 0.18843404808317088,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004999115302546866,
|
|
"loss": 6.9474,
|
|
"mean_token_accuracy": 0.12130062356591224,
|
|
"num_tokens": 3704806.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.132661199569702,
|
|
"epoch": 0.18893387314439947,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004999105267019486,
|
|
"loss": 7.0117,
|
|
"mean_token_accuracy": 0.1149711400270462,
|
|
"num_tokens": 3713927.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.325976610183716,
|
|
"epoch": 0.18943369820562803,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999095174905195,
|
|
"loss": 7.0869,
|
|
"mean_token_accuracy": 0.10872976779937744,
|
|
"num_tokens": 3723685.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.113046550750733,
|
|
"epoch": 0.1899335232668566,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999085026204249,
|
|
"loss": 7.0462,
|
|
"mean_token_accuracy": 0.11273184865713119,
|
|
"num_tokens": 3733918.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.19275393486023,
|
|
"epoch": 0.19043334832808517,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004999074820916903,
|
|
"loss": 7.0139,
|
|
"mean_token_accuracy": 0.10976849347352982,
|
|
"num_tokens": 3744466.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.280959558486939,
|
|
"epoch": 0.19093317338931373,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999064559043412,
|
|
"loss": 6.9644,
|
|
"mean_token_accuracy": 0.11463172510266303,
|
|
"num_tokens": 3754082.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.130083608627319,
|
|
"epoch": 0.19143299845054232,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999054240584037,
|
|
"loss": 6.9828,
|
|
"mean_token_accuracy": 0.11847582682967187,
|
|
"num_tokens": 3763111.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.268581295013428,
|
|
"epoch": 0.19193282351177088,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999043865539035,
|
|
"loss": 7.0493,
|
|
"mean_token_accuracy": 0.11117423921823502,
|
|
"num_tokens": 3772555.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.268717622756958,
|
|
"epoch": 0.19243264857299944,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999033433908667,
|
|
"loss": 7.0815,
|
|
"mean_token_accuracy": 0.111016134172678,
|
|
"num_tokens": 3783073.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 7.111140966415405,
|
|
"epoch": 0.19293247363422802,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999022945693198,
|
|
"loss": 7.0262,
|
|
"mean_token_accuracy": 0.11531205549836158,
|
|
"num_tokens": 3792277.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.230959272384643,
|
|
"epoch": 0.19343229869545658,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499901240089289,
|
|
"loss": 6.9675,
|
|
"mean_token_accuracy": 0.11661996394395828,
|
|
"num_tokens": 3801098.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 7.192171192169189,
|
|
"epoch": 0.19393212375668517,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999001799508009,
|
|
"loss": 7.0099,
|
|
"mean_token_accuracy": 0.11187705099582672,
|
|
"num_tokens": 3810614.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 7.157060289382935,
|
|
"epoch": 0.19443194881791373,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998991141538821,
|
|
"loss": 6.9944,
|
|
"mean_token_accuracy": 0.11400014832615853,
|
|
"num_tokens": 3818989.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 7.1906585693359375,
|
|
"epoch": 0.1949317738791423,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998980426985596,
|
|
"loss": 7.024,
|
|
"mean_token_accuracy": 0.11215802431106567,
|
|
"num_tokens": 3829319.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.190929889678955,
|
|
"epoch": 0.19543159894037088,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00049989696558486,
|
|
"loss": 7.0554,
|
|
"mean_token_accuracy": 0.11058950945734977,
|
|
"num_tokens": 3839545.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.1641429424285885,
|
|
"epoch": 0.19593142400159944,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998958828128108,
|
|
"loss": 6.9876,
|
|
"mean_token_accuracy": 0.11402552798390389,
|
|
"num_tokens": 3849029.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.209018182754517,
|
|
"epoch": 0.19643124906282802,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998947943824389,
|
|
"loss": 7.0268,
|
|
"mean_token_accuracy": 0.11167576164007187,
|
|
"num_tokens": 3859463.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.266246891021728,
|
|
"epoch": 0.19693107412405658,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998937002937719,
|
|
"loss": 7.072,
|
|
"mean_token_accuracy": 0.1130496896803379,
|
|
"num_tokens": 3869695.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 7.197639513015747,
|
|
"epoch": 0.19743089918528514,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004998926005468373,
|
|
"loss": 7.0303,
|
|
"mean_token_accuracy": 0.11531598567962646,
|
|
"num_tokens": 3879608.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.166806554794311,
|
|
"epoch": 0.19793072424651373,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998914951416627,
|
|
"loss": 6.942,
|
|
"mean_token_accuracy": 0.11460894122719764,
|
|
"num_tokens": 3888709.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 7.064781713485718,
|
|
"epoch": 0.1984305493077423,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499890384078276,
|
|
"loss": 6.9031,
|
|
"mean_token_accuracy": 0.11336119174957275,
|
|
"num_tokens": 3898392.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 7.218285322189331,
|
|
"epoch": 0.19893037436897085,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998892673567051,
|
|
"loss": 7.1152,
|
|
"mean_token_accuracy": 0.1025274470448494,
|
|
"num_tokens": 3909508.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 7.13175253868103,
|
|
"epoch": 0.19943019943019943,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004998881449769781,
|
|
"loss": 6.9752,
|
|
"mean_token_accuracy": 0.11452259048819542,
|
|
"num_tokens": 3918155.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 7.231250333786011,
|
|
"epoch": 0.199930024491428,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998870169391232,
|
|
"loss": 7.0059,
|
|
"mean_token_accuracy": 0.11461703926324844,
|
|
"num_tokens": 3928478.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 7.109054040908814,
|
|
"epoch": 0.20042984955265658,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499885883243169,
|
|
"loss": 7.0509,
|
|
"mean_token_accuracy": 0.11921404525637627,
|
|
"num_tokens": 3938126.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 7.078082656860351,
|
|
"epoch": 0.20092967461388514,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998847438891437,
|
|
"loss": 6.8813,
|
|
"mean_token_accuracy": 0.1193962648510933,
|
|
"num_tokens": 3948176.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 7.241360139846802,
|
|
"epoch": 0.2014294996751137,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998835988770761,
|
|
"loss": 6.9992,
|
|
"mean_token_accuracy": 0.11483975872397423,
|
|
"num_tokens": 3958983.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 7.210058927536011,
|
|
"epoch": 0.20192932473634229,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000499882448206995,
|
|
"loss": 6.9562,
|
|
"mean_token_accuracy": 0.11502718180418015,
|
|
"num_tokens": 3968766.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 7.144572973251343,
|
|
"epoch": 0.20242914979757085,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998812918789293,
|
|
"loss": 6.9732,
|
|
"mean_token_accuracy": 0.11400391906499863,
|
|
"num_tokens": 3979487.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 7.132036590576172,
|
|
"epoch": 0.20292897485879943,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998801298929084,
|
|
"loss": 6.948,
|
|
"mean_token_accuracy": 0.11947876140475273,
|
|
"num_tokens": 3988450.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 7.156334114074707,
|
|
"epoch": 0.203428799920028,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998789622489611,
|
|
"loss": 6.9566,
|
|
"mean_token_accuracy": 0.11755107268691063,
|
|
"num_tokens": 3997492.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 7.110681772232056,
|
|
"epoch": 0.20392862498125655,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998777889471171,
|
|
"loss": 6.9394,
|
|
"mean_token_accuracy": 0.12227256521582604,
|
|
"num_tokens": 4007298.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 7.190283393859863,
|
|
"epoch": 0.20442845004248514,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998766099874056,
|
|
"loss": 6.9835,
|
|
"mean_token_accuracy": 0.1158505879342556,
|
|
"num_tokens": 4017727.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 7.130951261520385,
|
|
"epoch": 0.2049282751037137,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998754253698566,
|
|
"loss": 6.979,
|
|
"mean_token_accuracy": 0.11403572708368301,
|
|
"num_tokens": 4027618.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 7.089246463775635,
|
|
"epoch": 0.20542810016494226,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004998742350944998,
|
|
"loss": 7.0016,
|
|
"mean_token_accuracy": 0.11411673203110695,
|
|
"num_tokens": 4038831.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 7.18756742477417,
|
|
"epoch": 0.20592792522617084,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004998730391613651,
|
|
"loss": 6.9477,
|
|
"mean_token_accuracy": 0.1187774159014225,
|
|
"num_tokens": 4048736.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 7.13449182510376,
|
|
"epoch": 0.2064277502873994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998718375704825,
|
|
"loss": 6.8957,
|
|
"mean_token_accuracy": 0.11968778371810913,
|
|
"num_tokens": 4057633.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 7.130615329742431,
|
|
"epoch": 0.206927575348628,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998706303218825,
|
|
"loss": 7.011,
|
|
"mean_token_accuracy": 0.12041295915842057,
|
|
"num_tokens": 4067883.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 7.19430775642395,
|
|
"epoch": 0.20742740040985655,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998694174155952,
|
|
"loss": 6.8841,
|
|
"mean_token_accuracy": 0.12097653448581695,
|
|
"num_tokens": 4078147.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 7.091365003585816,
|
|
"epoch": 0.2079272254710851,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998681988516512,
|
|
"loss": 6.8675,
|
|
"mean_token_accuracy": 0.11910612732172013,
|
|
"num_tokens": 4087130.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 7.035022354125976,
|
|
"epoch": 0.2084270505323137,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998669746300811,
|
|
"loss": 6.9829,
|
|
"mean_token_accuracy": 0.1219471551477909,
|
|
"num_tokens": 4096840.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 7.219481992721557,
|
|
"epoch": 0.20892687559354226,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004998657447509159,
|
|
"loss": 6.9685,
|
|
"mean_token_accuracy": 0.11537734568119049,
|
|
"num_tokens": 4105842.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 7.076082324981689,
|
|
"epoch": 0.20942670065477084,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998645092141864,
|
|
"loss": 6.9031,
|
|
"mean_token_accuracy": 0.12017038688063622,
|
|
"num_tokens": 4114865.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 7.1540521621704105,
|
|
"epoch": 0.2099265257159994,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998632680199238,
|
|
"loss": 7.0211,
|
|
"mean_token_accuracy": 0.11174777895212173,
|
|
"num_tokens": 4124954.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 7.097948169708252,
|
|
"epoch": 0.21042635077722796,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998620211681591,
|
|
"loss": 6.9188,
|
|
"mean_token_accuracy": 0.12242434993386268,
|
|
"num_tokens": 4134091.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 7.188031768798828,
|
|
"epoch": 0.21092617583845655,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998607686589239,
|
|
"loss": 6.9732,
|
|
"mean_token_accuracy": 0.11344101205468178,
|
|
"num_tokens": 4143566.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 7.129412221908569,
|
|
"epoch": 0.2114260008996851,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998595104922496,
|
|
"loss": 6.9414,
|
|
"mean_token_accuracy": 0.11348962932825088,
|
|
"num_tokens": 4153053.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 7.159371280670166,
|
|
"epoch": 0.2119258259609137,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998582466681678,
|
|
"loss": 6.9182,
|
|
"mean_token_accuracy": 0.1212809957563877,
|
|
"num_tokens": 4162925.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 7.011229467391968,
|
|
"epoch": 0.21242565102214225,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998569771867105,
|
|
"loss": 6.8495,
|
|
"mean_token_accuracy": 0.12126544713974,
|
|
"num_tokens": 4172077.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 7.090507984161377,
|
|
"epoch": 0.2129254760833708,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998557020479095,
|
|
"loss": 6.9376,
|
|
"mean_token_accuracy": 0.11842376366257668,
|
|
"num_tokens": 4182134.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 7.2148134231567385,
|
|
"epoch": 0.2134253011445994,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004998544212517968,
|
|
"loss": 6.9533,
|
|
"mean_token_accuracy": 0.11436854675412178,
|
|
"num_tokens": 4191963.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 7.040346622467041,
|
|
"epoch": 0.21392512620582796,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998531347984047,
|
|
"loss": 6.9051,
|
|
"mean_token_accuracy": 0.1191865548491478,
|
|
"num_tokens": 4202180.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 7.15973048210144,
|
|
"epoch": 0.21442495126705652,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004998518426877656,
|
|
"loss": 6.9311,
|
|
"mean_token_accuracy": 0.11763255894184113,
|
|
"num_tokens": 4211874.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 7.0943811416625975,
|
|
"epoch": 0.2149247763282851,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004998505449199122,
|
|
"loss": 6.885,
|
|
"mean_token_accuracy": 0.11530448347330094,
|
|
"num_tokens": 4221610.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.956639623641967,
|
|
"epoch": 0.21542460138951366,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998492414948768,
|
|
"loss": 6.8427,
|
|
"mean_token_accuracy": 0.12943953052163124,
|
|
"num_tokens": 4230610.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 7.111695718765259,
|
|
"epoch": 0.21592442645074225,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004998479324126924,
|
|
"loss": 6.9587,
|
|
"mean_token_accuracy": 0.11384863331913948,
|
|
"num_tokens": 4241389.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 7.122995376586914,
|
|
"epoch": 0.2164242515119708,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998466176733918,
|
|
"loss": 6.9007,
|
|
"mean_token_accuracy": 0.11816137433052062,
|
|
"num_tokens": 4250617.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 7.05834994316101,
|
|
"epoch": 0.21692407657319937,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998452972770082,
|
|
"loss": 6.8789,
|
|
"mean_token_accuracy": 0.11939359828829765,
|
|
"num_tokens": 4260000.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 7.071297264099121,
|
|
"epoch": 0.21742390163442796,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998439712235747,
|
|
"loss": 6.8335,
|
|
"mean_token_accuracy": 0.12091447785496712,
|
|
"num_tokens": 4269751.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 7.092474317550659,
|
|
"epoch": 0.21792372669565652,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998426395131248,
|
|
"loss": 6.8873,
|
|
"mean_token_accuracy": 0.11750671342015266,
|
|
"num_tokens": 4279210.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 7.095043182373047,
|
|
"epoch": 0.2184235517568851,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998413021456919,
|
|
"loss": 6.8667,
|
|
"mean_token_accuracy": 0.1229426383972168,
|
|
"num_tokens": 4288573.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 7.082563734054565,
|
|
"epoch": 0.21892337681811366,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998399591213098,
|
|
"loss": 6.8855,
|
|
"mean_token_accuracy": 0.11777178645133972,
|
|
"num_tokens": 4298438.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 7.016525602340698,
|
|
"epoch": 0.21942320187934222,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998386104400122,
|
|
"loss": 6.8181,
|
|
"mean_token_accuracy": 0.12750059515237808,
|
|
"num_tokens": 4307174.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 7.152014636993409,
|
|
"epoch": 0.2199230269405708,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998372561018329,
|
|
"loss": 6.9784,
|
|
"mean_token_accuracy": 0.11399625465273858,
|
|
"num_tokens": 4317220.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 7.140341758728027,
|
|
"epoch": 0.22042285200179937,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998358961068062,
|
|
"loss": 6.9007,
|
|
"mean_token_accuracy": 0.12417919039726258,
|
|
"num_tokens": 4326496.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 7.062141704559326,
|
|
"epoch": 0.22092267706302793,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998345304549661,
|
|
"loss": 6.9008,
|
|
"mean_token_accuracy": 0.12094885706901551,
|
|
"num_tokens": 4335705.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 7.156447696685791,
|
|
"epoch": 0.22142250212425652,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998331591463471,
|
|
"loss": 6.9136,
|
|
"mean_token_accuracy": 0.12432578578591347,
|
|
"num_tokens": 4344711.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 7.058556938171387,
|
|
"epoch": 0.22192232718548507,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998317821809837,
|
|
"loss": 6.9247,
|
|
"mean_token_accuracy": 0.11726782843470573,
|
|
"num_tokens": 4355448.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 7.016918325424195,
|
|
"epoch": 0.22242215224671366,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998303995589104,
|
|
"loss": 6.8137,
|
|
"mean_token_accuracy": 0.12222934365272523,
|
|
"num_tokens": 4365104.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 7.023652172088623,
|
|
"epoch": 0.22292197730794222,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998290112801623,
|
|
"loss": 6.7861,
|
|
"mean_token_accuracy": 0.12437113225460053,
|
|
"num_tokens": 4375219.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 7.107285213470459,
|
|
"epoch": 0.22342180236917078,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499827617344774,
|
|
"loss": 6.9092,
|
|
"mean_token_accuracy": 0.11431368067860603,
|
|
"num_tokens": 4385294.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 7.0897664546966555,
|
|
"epoch": 0.22392162743039937,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998262177527807,
|
|
"loss": 6.8791,
|
|
"mean_token_accuracy": 0.11411814317107201,
|
|
"num_tokens": 4394305.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 7.091776275634766,
|
|
"epoch": 0.22442145249162793,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004998248125042177,
|
|
"loss": 6.8152,
|
|
"mean_token_accuracy": 0.12024855241179466,
|
|
"num_tokens": 4403732.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 7.080003881454468,
|
|
"epoch": 0.2249212775528565,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998234015991202,
|
|
"loss": 6.9697,
|
|
"mean_token_accuracy": 0.11582587286829948,
|
|
"num_tokens": 4413730.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 7.186346435546875,
|
|
"epoch": 0.22542110261408507,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998219850375238,
|
|
"loss": 6.9858,
|
|
"mean_token_accuracy": 0.11293879821896553,
|
|
"num_tokens": 4423051.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 7.135729503631592,
|
|
"epoch": 0.22592092767531363,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004998205628194641,
|
|
"loss": 6.9801,
|
|
"mean_token_accuracy": 0.11710119098424912,
|
|
"num_tokens": 4433363.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 7.126741695404053,
|
|
"epoch": 0.22642075273654222,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004998191349449768,
|
|
"loss": 6.9249,
|
|
"mean_token_accuracy": 0.11564962193369865,
|
|
"num_tokens": 4442984.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 7.115010404586792,
|
|
"epoch": 0.22692057779777078,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998177014140981,
|
|
"loss": 6.876,
|
|
"mean_token_accuracy": 0.11640706211328507,
|
|
"num_tokens": 4452364.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 7.1195290088653564,
|
|
"epoch": 0.22742040285899934,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998162622268637,
|
|
"loss": 6.8947,
|
|
"mean_token_accuracy": 0.11545171216130257,
|
|
"num_tokens": 4461904.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 7.064880800247193,
|
|
"epoch": 0.22792022792022792,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998148173833101,
|
|
"loss": 6.8496,
|
|
"mean_token_accuracy": 0.12019126564264297,
|
|
"num_tokens": 4470633.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 7.072769355773926,
|
|
"epoch": 0.22842005298145648,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998133668834735,
|
|
"loss": 6.8757,
|
|
"mean_token_accuracy": 0.11745303049683571,
|
|
"num_tokens": 4480673.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 7.083282518386841,
|
|
"epoch": 0.22891987804268507,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998119107273905,
|
|
"loss": 6.8558,
|
|
"mean_token_accuracy": 0.11386666968464851,
|
|
"num_tokens": 4489923.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 7.023565435409546,
|
|
"epoch": 0.22941970310391363,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998104489150975,
|
|
"loss": 6.8231,
|
|
"mean_token_accuracy": 0.11956952810287476,
|
|
"num_tokens": 4501007.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 7.002195882797241,
|
|
"epoch": 0.2299195281651422,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998089814466318,
|
|
"loss": 6.851,
|
|
"mean_token_accuracy": 0.11942060068249702,
|
|
"num_tokens": 4510460.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 7.076708078384399,
|
|
"epoch": 0.23041935322637078,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004998075083220295,
|
|
"loss": 6.9317,
|
|
"mean_token_accuracy": 0.11818542778491974,
|
|
"num_tokens": 4521026.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 7.057536029815674,
|
|
"epoch": 0.23091917828759934,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998060295413286,
|
|
"loss": 6.8402,
|
|
"mean_token_accuracy": 0.12432301193475723,
|
|
"num_tokens": 4530818.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.965789461135865,
|
|
"epoch": 0.23141900334882792,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998045451045656,
|
|
"loss": 6.8025,
|
|
"mean_token_accuracy": 0.12702578529715539,
|
|
"num_tokens": 4539512.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 7.114251947402954,
|
|
"epoch": 0.23191882841005648,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998030550117782,
|
|
"loss": 6.9636,
|
|
"mean_token_accuracy": 0.1128588818013668,
|
|
"num_tokens": 4549030.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 7.038267183303833,
|
|
"epoch": 0.23241865347128504,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004998015592630037,
|
|
"loss": 6.8147,
|
|
"mean_token_accuracy": 0.12493672892451287,
|
|
"num_tokens": 4559347.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 7.0613970279693605,
|
|
"epoch": 0.23291847853251363,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004998000578582798,
|
|
"loss": 6.8676,
|
|
"mean_token_accuracy": 0.11753461137413979,
|
|
"num_tokens": 4568721.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 7.050783443450928,
|
|
"epoch": 0.2334183035937422,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997985507976443,
|
|
"loss": 6.8684,
|
|
"mean_token_accuracy": 0.1190462976694107,
|
|
"num_tokens": 4578265.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 7.009930086135864,
|
|
"epoch": 0.23391812865497075,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997970380811351,
|
|
"loss": 6.8525,
|
|
"mean_token_accuracy": 0.12626971676945686,
|
|
"num_tokens": 4588694.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 7.092075347900391,
|
|
"epoch": 0.23441795371619933,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997955197087904,
|
|
"loss": 6.8315,
|
|
"mean_token_accuracy": 0.12784341126680374,
|
|
"num_tokens": 4597754.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 7.026864671707154,
|
|
"epoch": 0.2349177787774279,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997939956806481,
|
|
"loss": 6.8521,
|
|
"mean_token_accuracy": 0.12023111805319786,
|
|
"num_tokens": 4606874.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.982235383987427,
|
|
"epoch": 0.23541760383865648,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997924659967467,
|
|
"loss": 6.7391,
|
|
"mean_token_accuracy": 0.12449168413877487,
|
|
"num_tokens": 4615588.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.959120845794677,
|
|
"epoch": 0.23591742889988504,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997909306571248,
|
|
"loss": 6.8909,
|
|
"mean_token_accuracy": 0.12301145270466804,
|
|
"num_tokens": 4625919.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 7.058623456954956,
|
|
"epoch": 0.2364172539611136,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004997893896618209,
|
|
"loss": 6.8355,
|
|
"mean_token_accuracy": 0.11596318781375885,
|
|
"num_tokens": 4635539.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 7.008761596679688,
|
|
"epoch": 0.2369170790223422,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004997878430108738,
|
|
"loss": 6.8361,
|
|
"mean_token_accuracy": 0.11988401189446449,
|
|
"num_tokens": 4645833.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.989094305038452,
|
|
"epoch": 0.23741690408357075,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997862907043224,
|
|
"loss": 6.9397,
|
|
"mean_token_accuracy": 0.12021735459566116,
|
|
"num_tokens": 4655166.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 7.0768798828125,
|
|
"epoch": 0.23791672914479933,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997847327422059,
|
|
"loss": 6.8216,
|
|
"mean_token_accuracy": 0.11937851533293724,
|
|
"num_tokens": 4664884.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 7.013132905960083,
|
|
"epoch": 0.2384165542060279,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997831691245632,
|
|
"loss": 6.9188,
|
|
"mean_token_accuracy": 0.11891724541783333,
|
|
"num_tokens": 4675172.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 7.094194173812866,
|
|
"epoch": 0.23891637926725645,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997815998514337,
|
|
"loss": 6.8083,
|
|
"mean_token_accuracy": 0.12721233293414116,
|
|
"num_tokens": 4684588.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 7.038353013992309,
|
|
"epoch": 0.23941620432848504,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997800249228572,
|
|
"loss": 6.9273,
|
|
"mean_token_accuracy": 0.11809542998671532,
|
|
"num_tokens": 4694434.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 7.012913179397583,
|
|
"epoch": 0.2399160293897136,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004997784443388732,
|
|
"loss": 6.8729,
|
|
"mean_token_accuracy": 0.1202128566801548,
|
|
"num_tokens": 4705080.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 7.109264183044433,
|
|
"epoch": 0.24041585445094216,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004997768580995212,
|
|
"loss": 6.9413,
|
|
"mean_token_accuracy": 0.11637584567070007,
|
|
"num_tokens": 4714466.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 7.006923818588257,
|
|
"epoch": 0.24091567951217074,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004997752662048413,
|
|
"loss": 6.8346,
|
|
"mean_token_accuracy": 0.12132125496864318,
|
|
"num_tokens": 4724047.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 7.0618185043334964,
|
|
"epoch": 0.2414155045733993,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997736686548736,
|
|
"loss": 6.8815,
|
|
"mean_token_accuracy": 0.11511459723114967,
|
|
"num_tokens": 4733153.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 7.07546854019165,
|
|
"epoch": 0.2419153296346279,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004997720654496582,
|
|
"loss": 6.7452,
|
|
"mean_token_accuracy": 0.12942711487412453,
|
|
"num_tokens": 4743239.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.986814689636231,
|
|
"epoch": 0.24241515469585645,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997704565892355,
|
|
"loss": 6.8139,
|
|
"mean_token_accuracy": 0.12030862644314766,
|
|
"num_tokens": 4753128.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 7.173285341262817,
|
|
"epoch": 0.242914979757085,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997688420736459,
|
|
"loss": 7.0347,
|
|
"mean_token_accuracy": 0.1086900345981121,
|
|
"num_tokens": 4764056.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 7.03861436843872,
|
|
"epoch": 0.2434148048183136,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997672219029301,
|
|
"loss": 6.8678,
|
|
"mean_token_accuracy": 0.11879493892192841,
|
|
"num_tokens": 4773354.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.988547229766846,
|
|
"epoch": 0.24391462987954216,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997655960771288,
|
|
"loss": 6.8062,
|
|
"mean_token_accuracy": 0.11831172928214073,
|
|
"num_tokens": 4783367.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.978290653228759,
|
|
"epoch": 0.24441445494077074,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000499763964596283,
|
|
"loss": 6.7079,
|
|
"mean_token_accuracy": 0.1302500769495964,
|
|
"num_tokens": 4793028.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 7.027144050598144,
|
|
"epoch": 0.2449142800019993,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997623274604336,
|
|
"loss": 6.8748,
|
|
"mean_token_accuracy": 0.11989556774497032,
|
|
"num_tokens": 4804032.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 7.083219337463379,
|
|
"epoch": 0.24541410506322786,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499760684669622,
|
|
"loss": 6.9002,
|
|
"mean_token_accuracy": 0.12228977680206299,
|
|
"num_tokens": 4813294.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.98912935256958,
|
|
"epoch": 0.24591393012445645,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997590362238894,
|
|
"loss": 6.807,
|
|
"mean_token_accuracy": 0.12156101688742638,
|
|
"num_tokens": 4823461.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.951521015167236,
|
|
"epoch": 0.246413755185685,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004997573821232772,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.12493078187108039,
|
|
"num_tokens": 4832464.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 7.01842360496521,
|
|
"epoch": 0.24691358024691357,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997557223678272,
|
|
"loss": 6.8186,
|
|
"mean_token_accuracy": 0.12038722783327102,
|
|
"num_tokens": 4842697.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 7.077435159683228,
|
|
"epoch": 0.24741340530814215,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997540569575809,
|
|
"loss": 6.8702,
|
|
"mean_token_accuracy": 0.12191974818706512,
|
|
"num_tokens": 4853129.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 7.058256053924561,
|
|
"epoch": 0.2479132303693707,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004997523858925805,
|
|
"loss": 6.908,
|
|
"mean_token_accuracy": 0.11191392317414284,
|
|
"num_tokens": 4864048.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.903012084960937,
|
|
"epoch": 0.2484130554305993,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997507091728678,
|
|
"loss": 6.8637,
|
|
"mean_token_accuracy": 0.12203898876905442,
|
|
"num_tokens": 4874368.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 7.099834775924682,
|
|
"epoch": 0.24891288049182786,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004997490267984853,
|
|
"loss": 6.8287,
|
|
"mean_token_accuracy": 0.12041815221309662,
|
|
"num_tokens": 4884813.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 7.00856614112854,
|
|
"epoch": 0.24941270555305642,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997473387694749,
|
|
"loss": 6.8915,
|
|
"mean_token_accuracy": 0.1209457166492939,
|
|
"num_tokens": 4895594.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.982333946228027,
|
|
"epoch": 0.249912530614285,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997456450858793,
|
|
"loss": 6.781,
|
|
"mean_token_accuracy": 0.12502407133579255,
|
|
"num_tokens": 4904922.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 7.018849563598633,
|
|
"epoch": 0.25041235567551356,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997439457477412,
|
|
"loss": 6.7902,
|
|
"mean_token_accuracy": 0.12333055436611176,
|
|
"num_tokens": 4915009.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 7.075725412368774,
|
|
"epoch": 0.25091218073674215,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997422407551032,
|
|
"loss": 6.918,
|
|
"mean_token_accuracy": 0.11689607426524162,
|
|
"num_tokens": 4924916.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 7.0010294914245605,
|
|
"epoch": 0.2514120057979707,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004997405301080082,
|
|
"loss": 6.8258,
|
|
"mean_token_accuracy": 0.11776653751730919,
|
|
"num_tokens": 4935127.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 7.039859199523926,
|
|
"epoch": 0.25191183085919927,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997388138064994,
|
|
"loss": 6.8401,
|
|
"mean_token_accuracy": 0.12345104441046714,
|
|
"num_tokens": 4944749.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.964186906814575,
|
|
"epoch": 0.25241165592042786,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997370918506198,
|
|
"loss": 6.7002,
|
|
"mean_token_accuracy": 0.1314247652888298,
|
|
"num_tokens": 4953868.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.972427940368652,
|
|
"epoch": 0.25291148098165644,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004997353642404129,
|
|
"loss": 6.7993,
|
|
"mean_token_accuracy": 0.12143997326493264,
|
|
"num_tokens": 4963282.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.901938772201538,
|
|
"epoch": 0.253411306042885,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997336309759219,
|
|
"loss": 6.6345,
|
|
"mean_token_accuracy": 0.13661187663674354,
|
|
"num_tokens": 4973099.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 7.017197799682617,
|
|
"epoch": 0.25391113110411356,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997318920571907,
|
|
"loss": 6.8116,
|
|
"mean_token_accuracy": 0.11694255024194718,
|
|
"num_tokens": 4983771.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.881151914596558,
|
|
"epoch": 0.25441095616534215,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499730147484263,
|
|
"loss": 6.6461,
|
|
"mean_token_accuracy": 0.12338883578777313,
|
|
"num_tokens": 4993648.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.960689496994019,
|
|
"epoch": 0.2549107812265707,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997283972571824,
|
|
"loss": 6.8234,
|
|
"mean_token_accuracy": 0.12306213602423668,
|
|
"num_tokens": 5002821.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 7.070871496200562,
|
|
"epoch": 0.25541060628779927,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997266413759934,
|
|
"loss": 6.8332,
|
|
"mean_token_accuracy": 0.12124887555837631,
|
|
"num_tokens": 5012655.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.9131255626678465,
|
|
"epoch": 0.25591043134902786,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004997248798407397,
|
|
"loss": 6.8003,
|
|
"mean_token_accuracy": 0.12324445545673371,
|
|
"num_tokens": 5023649.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.994232654571533,
|
|
"epoch": 0.2564102564102564,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997231126514659,
|
|
"loss": 6.754,
|
|
"mean_token_accuracy": 0.1276788279414177,
|
|
"num_tokens": 5031945.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 7.038708591461182,
|
|
"epoch": 0.256910081471485,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0004997213398082165,
|
|
"loss": 6.8319,
|
|
"mean_token_accuracy": 0.11983529329299927,
|
|
"num_tokens": 5042036.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.972916555404663,
|
|
"epoch": 0.25740990653271356,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000499719561311036,
|
|
"loss": 6.8136,
|
|
"mean_token_accuracy": 0.11984457820653915,
|
|
"num_tokens": 5050986.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.988238525390625,
|
|
"epoch": 0.2579097315939421,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997177771599691,
|
|
"loss": 6.7824,
|
|
"mean_token_accuracy": 0.12704542875289918,
|
|
"num_tokens": 5061408.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.902217388153076,
|
|
"epoch": 0.2584095566551707,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997159873550607,
|
|
"loss": 6.7192,
|
|
"mean_token_accuracy": 0.12395873069763183,
|
|
"num_tokens": 5071646.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.9063128471374515,
|
|
"epoch": 0.25890938171639927,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004997141918963561,
|
|
"loss": 6.8173,
|
|
"mean_token_accuracy": 0.1257748030126095,
|
|
"num_tokens": 5083185.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 7.0397814273834225,
|
|
"epoch": 0.25940920677762785,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997123907839001,
|
|
"loss": 6.7881,
|
|
"mean_token_accuracy": 0.12251957878470421,
|
|
"num_tokens": 5092993.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.979730272293091,
|
|
"epoch": 0.2599090318388564,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997105840177381,
|
|
"loss": 6.8354,
|
|
"mean_token_accuracy": 0.117799261957407,
|
|
"num_tokens": 5102709.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 7.008151340484619,
|
|
"epoch": 0.260408856900085,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997087715979157,
|
|
"loss": 6.7792,
|
|
"mean_token_accuracy": 0.11763950660824776,
|
|
"num_tokens": 5112858.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 7.046616935729981,
|
|
"epoch": 0.26090868196131356,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004997069535244785,
|
|
"loss": 6.6954,
|
|
"mean_token_accuracy": 0.12784051820635794,
|
|
"num_tokens": 5123573.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.8589095115661625,
|
|
"epoch": 0.2614085070225421,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997051297974722,
|
|
"loss": 6.8639,
|
|
"mean_token_accuracy": 0.11722040623426437,
|
|
"num_tokens": 5132908.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 7.009056949615479,
|
|
"epoch": 0.2619083320837707,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004997033004169425,
|
|
"loss": 6.755,
|
|
"mean_token_accuracy": 0.1255860634148121,
|
|
"num_tokens": 5142331.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 7.080911350250244,
|
|
"epoch": 0.26240815714499927,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004997014653829357,
|
|
"loss": 6.8289,
|
|
"mean_token_accuracy": 0.1180228516459465,
|
|
"num_tokens": 5152517.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.921509647369385,
|
|
"epoch": 0.2629079822062278,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004996996246954977,
|
|
"loss": 6.7802,
|
|
"mean_token_accuracy": 0.12336174324154854,
|
|
"num_tokens": 5162418.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.944567775726318,
|
|
"epoch": 0.2634078072674564,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004996977783546752,
|
|
"loss": 6.797,
|
|
"mean_token_accuracy": 0.12096343412995339,
|
|
"num_tokens": 5171948.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 7.008423233032227,
|
|
"epoch": 0.26390763232868497,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996959263605143,
|
|
"loss": 6.7226,
|
|
"mean_token_accuracy": 0.12412903606891632,
|
|
"num_tokens": 5180963.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.923210000991821,
|
|
"epoch": 0.2644074573899135,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004996940687130617,
|
|
"loss": 6.8323,
|
|
"mean_token_accuracy": 0.12230520248413086,
|
|
"num_tokens": 5192305.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 7.007975339889526,
|
|
"epoch": 0.2649072824511421,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996922054123641,
|
|
"loss": 6.7307,
|
|
"mean_token_accuracy": 0.12070221900939941,
|
|
"num_tokens": 5201503.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.982429170608521,
|
|
"epoch": 0.2654071075123707,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996903364584685,
|
|
"loss": 6.6724,
|
|
"mean_token_accuracy": 0.1253625251352787,
|
|
"num_tokens": 5210206.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.814099502563477,
|
|
"epoch": 0.26590693257359926,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004996884618514219,
|
|
"loss": 6.7081,
|
|
"mean_token_accuracy": 0.12672959715127946,
|
|
"num_tokens": 5220256.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.92271785736084,
|
|
"epoch": 0.2664067576348278,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996865815912713,
|
|
"loss": 6.7307,
|
|
"mean_token_accuracy": 0.12912995740771294,
|
|
"num_tokens": 5229128.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.972404956817627,
|
|
"epoch": 0.2669065826960564,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996846956780642,
|
|
"loss": 6.6754,
|
|
"mean_token_accuracy": 0.12550344914197922,
|
|
"num_tokens": 5239468.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.915175199508667,
|
|
"epoch": 0.26740640775728497,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000499682804111848,
|
|
"loss": 6.769,
|
|
"mean_token_accuracy": 0.12083193883299828,
|
|
"num_tokens": 5247957.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.947061824798584,
|
|
"epoch": 0.2679062328185135,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004996809068926702,
|
|
"loss": 6.7875,
|
|
"mean_token_accuracy": 0.11905844509601593,
|
|
"num_tokens": 5257960.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.928300905227661,
|
|
"epoch": 0.2684060578797421,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004996790040205786,
|
|
"loss": 6.6633,
|
|
"mean_token_accuracy": 0.1272826075553894,
|
|
"num_tokens": 5266696.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.9380433559417725,
|
|
"epoch": 0.2689058829409707,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996770954956211,
|
|
"loss": 6.6382,
|
|
"mean_token_accuracy": 0.1306777834892273,
|
|
"num_tokens": 5275619.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.8965309143066404,
|
|
"epoch": 0.2694057080021992,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004996751813178456,
|
|
"loss": 6.7897,
|
|
"mean_token_accuracy": 0.12241848036646844,
|
|
"num_tokens": 5284884.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.951219177246093,
|
|
"epoch": 0.2699055330634278,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996732614873005,
|
|
"loss": 6.6823,
|
|
"mean_token_accuracy": 0.13157505616545678,
|
|
"num_tokens": 5293909.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.885016298294067,
|
|
"epoch": 0.2704053581246564,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996713360040339,
|
|
"loss": 6.7472,
|
|
"mean_token_accuracy": 0.11952828168869019,
|
|
"num_tokens": 5303142.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.935619926452636,
|
|
"epoch": 0.27090518318588497,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004996694048680942,
|
|
"loss": 6.6925,
|
|
"mean_token_accuracy": 0.12288326397538185,
|
|
"num_tokens": 5313055.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.9312550067901615,
|
|
"epoch": 0.2714050082471135,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004996674680795302,
|
|
"loss": 6.8109,
|
|
"mean_token_accuracy": 0.11529392302036286,
|
|
"num_tokens": 5323773.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.812974977493286,
|
|
"epoch": 0.2719048333083421,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004996655256383905,
|
|
"loss": 6.5991,
|
|
"mean_token_accuracy": 0.13647473528981208,
|
|
"num_tokens": 5334613.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.891908311843872,
|
|
"epoch": 0.2724046583695707,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000499663577544724,
|
|
"loss": 6.6816,
|
|
"mean_token_accuracy": 0.12988344877958297,
|
|
"num_tokens": 5344151.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.891177415847778,
|
|
"epoch": 0.2729044834307992,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004996616237985796,
|
|
"loss": 6.6389,
|
|
"mean_token_accuracy": 0.13222801834344863,
|
|
"num_tokens": 5352425.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.880789041519165,
|
|
"epoch": 0.2734043084920278,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004996596644000066,
|
|
"loss": 6.8408,
|
|
"mean_token_accuracy": 0.1188066765666008,
|
|
"num_tokens": 5361811.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.9263018608093265,
|
|
"epoch": 0.2739041335532564,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004996576993490543,
|
|
"loss": 6.786,
|
|
"mean_token_accuracy": 0.11447468400001526,
|
|
"num_tokens": 5371463.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.908780908584594,
|
|
"epoch": 0.2744039586144849,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004996557286457721,
|
|
"loss": 6.6502,
|
|
"mean_token_accuracy": 0.12994811981916427,
|
|
"num_tokens": 5381464.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.861965656280518,
|
|
"epoch": 0.2749037836757135,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996537522902096,
|
|
"loss": 6.7194,
|
|
"mean_token_accuracy": 0.12399204149842262,
|
|
"num_tokens": 5391012.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.97508602142334,
|
|
"epoch": 0.2754036087369421,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004996517702824163,
|
|
"loss": 6.8309,
|
|
"mean_token_accuracy": 0.12259437888860703,
|
|
"num_tokens": 5401198.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 7.017956781387329,
|
|
"epoch": 0.2759034337981706,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004996497826224425,
|
|
"loss": 6.7654,
|
|
"mean_token_accuracy": 0.12429968118667603,
|
|
"num_tokens": 5410592.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.83931188583374,
|
|
"epoch": 0.2764032588593992,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996477893103379,
|
|
"loss": 6.7505,
|
|
"mean_token_accuracy": 0.1281308613717556,
|
|
"num_tokens": 5420878.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.955085563659668,
|
|
"epoch": 0.2769030839206278,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996457903461527,
|
|
"loss": 6.7523,
|
|
"mean_token_accuracy": 0.12567364796996117,
|
|
"num_tokens": 5431199.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.932239675521851,
|
|
"epoch": 0.2774029089818564,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004996437857299373,
|
|
"loss": 6.6819,
|
|
"mean_token_accuracy": 0.12451341077685356,
|
|
"num_tokens": 5441661.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.778904724121094,
|
|
"epoch": 0.2779027340430849,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000499641775461742,
|
|
"loss": 6.6714,
|
|
"mean_token_accuracy": 0.132535882294178,
|
|
"num_tokens": 5451863.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.936461877822876,
|
|
"epoch": 0.2784025591043135,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996397595416173,
|
|
"loss": 6.7345,
|
|
"mean_token_accuracy": 0.12622895315289498,
|
|
"num_tokens": 5461600.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 7.027527093887329,
|
|
"epoch": 0.2789023841655421,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996377379696143,
|
|
"loss": 6.7576,
|
|
"mean_token_accuracy": 0.12218005806207657,
|
|
"num_tokens": 5470097.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.948017263412476,
|
|
"epoch": 0.2794022092267706,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996357107457834,
|
|
"loss": 6.8869,
|
|
"mean_token_accuracy": 0.11402522325515747,
|
|
"num_tokens": 5480906.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 7.057219552993774,
|
|
"epoch": 0.2799020342879992,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499633677870176,
|
|
"loss": 6.7821,
|
|
"mean_token_accuracy": 0.11816998645663261,
|
|
"num_tokens": 5489732.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.840402030944825,
|
|
"epoch": 0.2804018593492278,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004996316393428429,
|
|
"loss": 6.602,
|
|
"mean_token_accuracy": 0.13391595482826232,
|
|
"num_tokens": 5498422.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.822567796707153,
|
|
"epoch": 0.2809016844104563,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004996295951638357,
|
|
"loss": 6.7189,
|
|
"mean_token_accuracy": 0.12329033613204957,
|
|
"num_tokens": 5507958.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.983218050003051,
|
|
"epoch": 0.2814015094716849,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996275453332057,
|
|
"loss": 6.761,
|
|
"mean_token_accuracy": 0.12427213788032532,
|
|
"num_tokens": 5516752.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.899249696731568,
|
|
"epoch": 0.2819013345329135,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004996254898510043,
|
|
"loss": 6.7768,
|
|
"mean_token_accuracy": 0.12306348532438278,
|
|
"num_tokens": 5527211.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.921328783035278,
|
|
"epoch": 0.282401159594142,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996234287172835,
|
|
"loss": 6.6782,
|
|
"mean_token_accuracy": 0.12412735596299171,
|
|
"num_tokens": 5536925.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.966558170318604,
|
|
"epoch": 0.2829009846553706,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499621361932095,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.11656751185655594,
|
|
"num_tokens": 5546800.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.868718194961548,
|
|
"epoch": 0.2834008097165992,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996192894954909,
|
|
"loss": 6.697,
|
|
"mean_token_accuracy": 0.1250438630580902,
|
|
"num_tokens": 5557266.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.938524961471558,
|
|
"epoch": 0.2839006347778278,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996172114075232,
|
|
"loss": 6.6673,
|
|
"mean_token_accuracy": 0.12535413205623627,
|
|
"num_tokens": 5566169.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.896262836456299,
|
|
"epoch": 0.2844004598390563,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996151276682443,
|
|
"loss": 6.7617,
|
|
"mean_token_accuracy": 0.12486017644405364,
|
|
"num_tokens": 5575566.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.885776615142822,
|
|
"epoch": 0.2849002849002849,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004996130382777065,
|
|
"loss": 6.647,
|
|
"mean_token_accuracy": 0.12700871676206588,
|
|
"num_tokens": 5584804.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.893292379379273,
|
|
"epoch": 0.2854001099615135,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004996109432359625,
|
|
"loss": 6.7104,
|
|
"mean_token_accuracy": 0.12000066190958023,
|
|
"num_tokens": 5594361.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 7.01403489112854,
|
|
"epoch": 0.285899935022742,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0004996088425430651,
|
|
"loss": 6.6936,
|
|
"mean_token_accuracy": 0.1254715770483017,
|
|
"num_tokens": 5605682.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.884680128097534,
|
|
"epoch": 0.2863997600839706,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996067361990669,
|
|
"loss": 6.7746,
|
|
"mean_token_accuracy": 0.11833036169409752,
|
|
"num_tokens": 5617251.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.8461815357208256,
|
|
"epoch": 0.2868995851451992,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004996046242040209,
|
|
"loss": 6.6684,
|
|
"mean_token_accuracy": 0.1276953287422657,
|
|
"num_tokens": 5627140.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.882718229293824,
|
|
"epoch": 0.28739941020642773,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996025065579805,
|
|
"loss": 6.5987,
|
|
"mean_token_accuracy": 0.13200223743915557,
|
|
"num_tokens": 5635486.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.946580982208252,
|
|
"epoch": 0.2878992352676563,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004996003832609988,
|
|
"loss": 6.7686,
|
|
"mean_token_accuracy": 0.11976373195648193,
|
|
"num_tokens": 5645621.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.974322843551636,
|
|
"epoch": 0.2883990603288849,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995982543131292,
|
|
"loss": 6.8311,
|
|
"mean_token_accuracy": 0.11587899774312974,
|
|
"num_tokens": 5655626.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.923132610321045,
|
|
"epoch": 0.28889888539011344,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004995961197144254,
|
|
"loss": 6.7721,
|
|
"mean_token_accuracy": 0.12438612058758736,
|
|
"num_tokens": 5664547.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.959670162200927,
|
|
"epoch": 0.289398710451342,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499593979464941,
|
|
"loss": 6.7344,
|
|
"mean_token_accuracy": 0.12106872946023942,
|
|
"num_tokens": 5674861.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.850112104415894,
|
|
"epoch": 0.2898985355125706,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004995918335647298,
|
|
"loss": 6.6879,
|
|
"mean_token_accuracy": 0.12462930828332901,
|
|
"num_tokens": 5684143.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.901785516738892,
|
|
"epoch": 0.2903983605737992,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004995896820138459,
|
|
"loss": 6.6542,
|
|
"mean_token_accuracy": 0.12453328371047974,
|
|
"num_tokens": 5693837.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.979707384109497,
|
|
"epoch": 0.29089818563502773,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004995875248123434,
|
|
"loss": 6.7457,
|
|
"mean_token_accuracy": 0.11954636201262474,
|
|
"num_tokens": 5704827.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.934035778045654,
|
|
"epoch": 0.2913980106962563,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004995853619602766,
|
|
"loss": 6.7828,
|
|
"mean_token_accuracy": 0.12128729447722435,
|
|
"num_tokens": 5714419.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.87302975654602,
|
|
"epoch": 0.2918978357574849,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004995831934576998,
|
|
"loss": 6.7059,
|
|
"mean_token_accuracy": 0.12609893232584,
|
|
"num_tokens": 5724983.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.862426519393921,
|
|
"epoch": 0.29239766081871343,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004995810193046677,
|
|
"loss": 6.6389,
|
|
"mean_token_accuracy": 0.13176713660359382,
|
|
"num_tokens": 5734471.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.90787901878357,
|
|
"epoch": 0.292897485879942,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499578839501235,
|
|
"loss": 6.7643,
|
|
"mean_token_accuracy": 0.12451120018959046,
|
|
"num_tokens": 5744801.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.892666482925415,
|
|
"epoch": 0.2933973109411706,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004995766540474564,
|
|
"loss": 6.7164,
|
|
"mean_token_accuracy": 0.12045061364769935,
|
|
"num_tokens": 5754267.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.792302894592285,
|
|
"epoch": 0.29389713600239914,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000499574462943387,
|
|
"loss": 6.6261,
|
|
"mean_token_accuracy": 0.13216007202863694,
|
|
"num_tokens": 5764466.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.894134426116944,
|
|
"epoch": 0.2943969610636277,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000499572266189082,
|
|
"loss": 6.7673,
|
|
"mean_token_accuracy": 0.11944599896669388,
|
|
"num_tokens": 5775158.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.95402045249939,
|
|
"epoch": 0.2948967861248563,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004995700637845964,
|
|
"loss": 6.5789,
|
|
"mean_token_accuracy": 0.13158607333898545,
|
|
"num_tokens": 5783320.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.822439527511596,
|
|
"epoch": 0.29539661118608485,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004995678557299859,
|
|
"loss": 6.7291,
|
|
"mean_token_accuracy": 0.12602196037769317,
|
|
"num_tokens": 5792980.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.934293985366821,
|
|
"epoch": 0.29589643624731343,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004995656420253058,
|
|
"loss": 6.7596,
|
|
"mean_token_accuracy": 0.12525345459580423,
|
|
"num_tokens": 5802338.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.895787000656128,
|
|
"epoch": 0.296396261308542,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499563422670612,
|
|
"loss": 6.6529,
|
|
"mean_token_accuracy": 0.12653236985206603,
|
|
"num_tokens": 5811388.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.873412609100342,
|
|
"epoch": 0.2968960863697706,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004995611976659604,
|
|
"loss": 6.6797,
|
|
"mean_token_accuracy": 0.12875316813588142,
|
|
"num_tokens": 5821108.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.890844678878784,
|
|
"epoch": 0.29739591143099914,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004995589670114067,
|
|
"loss": 6.7279,
|
|
"mean_token_accuracy": 0.11901220306754112,
|
|
"num_tokens": 5831389.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.821516752243042,
|
|
"epoch": 0.2978957364922277,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004995567307070072,
|
|
"loss": 6.6059,
|
|
"mean_token_accuracy": 0.13066981732845306,
|
|
"num_tokens": 5840392.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.8590656280517575,
|
|
"epoch": 0.2983955615534563,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004995544887528182,
|
|
"loss": 6.6082,
|
|
"mean_token_accuracy": 0.12876412868499756,
|
|
"num_tokens": 5850032.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.887386322021484,
|
|
"epoch": 0.29889538661468484,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995522411488959,
|
|
"loss": 6.683,
|
|
"mean_token_accuracy": 0.12640040144324302,
|
|
"num_tokens": 5859204.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.845551586151123,
|
|
"epoch": 0.29939521167591343,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004995499878952971,
|
|
"loss": 6.6866,
|
|
"mean_token_accuracy": 0.1273339122533798,
|
|
"num_tokens": 5868582.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.876190519332885,
|
|
"epoch": 0.299895036737142,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004995477289920784,
|
|
"loss": 6.6229,
|
|
"mean_token_accuracy": 0.13258609995245935,
|
|
"num_tokens": 5879601.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.299895036737142,
|
|
"eval_entropy": 6.681353775320272,
|
|
"eval_loss": 6.711971759796143,
|
|
"eval_mean_token_accuracy": 0.13082575049137093,
|
|
"eval_num_tokens": 5879601.0,
|
|
"eval_runtime": 18.5417,
|
|
"eval_samples_per_second": 1674.337,
|
|
"eval_steps_per_second": 209.312,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 100030,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1391279704473600.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|