bd27151216
Model: fpadovani/ita-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
8123 lines
221 KiB
JSON
8123 lines
221 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.032274331820474,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.801114463806153,
|
|
"epoch": 0.005042864346949067,
|
|
"grad_norm": 14.3125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.3768,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11122.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.812651681900024,
|
|
"epoch": 0.010085728693898134,
|
|
"grad_norm": 15.5,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.2394,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20499.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.861662435531616,
|
|
"epoch": 0.015128593040847202,
|
|
"grad_norm": 20.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.0486,
|
|
"mean_token_accuracy": 0.00012091898825019597,
|
|
"num_tokens": 30664.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.152417182922363,
|
|
"epoch": 0.020171457387796268,
|
|
"grad_norm": 32.25,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.3871,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 41767.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.587842893600464,
|
|
"epoch": 0.025214321734745335,
|
|
"grad_norm": 14.25,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.6896,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 53233.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.463200283050536,
|
|
"epoch": 0.030257186081694403,
|
|
"grad_norm": 3.828125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.7364,
|
|
"mean_token_accuracy": 0.004124969011172652,
|
|
"num_tokens": 63591.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.693463897705078,
|
|
"epoch": 0.03530005042864347,
|
|
"grad_norm": 3.296875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.5492,
|
|
"mean_token_accuracy": 0.013820098433643579,
|
|
"num_tokens": 75211.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.683560562133788,
|
|
"epoch": 0.040342914775592535,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2263,
|
|
"mean_token_accuracy": 0.015752001013606787,
|
|
"num_tokens": 86163.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.606955909729004,
|
|
"epoch": 0.0453857791225416,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.9408,
|
|
"mean_token_accuracy": 0.02095262976363301,
|
|
"num_tokens": 96812.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.570794868469239,
|
|
"epoch": 0.05042864346949067,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.767,
|
|
"mean_token_accuracy": 0.03581012934446335,
|
|
"num_tokens": 107246.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.550295162200928,
|
|
"epoch": 0.05547150781643974,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.6982,
|
|
"mean_token_accuracy": 0.035054446384310725,
|
|
"num_tokens": 118169.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.557003593444824,
|
|
"epoch": 0.060514372163388806,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.596,
|
|
"mean_token_accuracy": 0.03818601593375206,
|
|
"num_tokens": 129744.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.555067825317384,
|
|
"epoch": 0.06555723651033787,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4879,
|
|
"mean_token_accuracy": 0.030292158760130406,
|
|
"num_tokens": 139826.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.539035034179687,
|
|
"epoch": 0.07060010085728693,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4026,
|
|
"mean_token_accuracy": 0.029551995545625688,
|
|
"num_tokens": 149767.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.505377006530761,
|
|
"epoch": 0.07564296520423601,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3373,
|
|
"mean_token_accuracy": 0.039892033115029334,
|
|
"num_tokens": 160895.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.469569301605224,
|
|
"epoch": 0.08068582955118507,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2884,
|
|
"mean_token_accuracy": 0.03964783567935228,
|
|
"num_tokens": 172080.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.455696964263916,
|
|
"epoch": 0.08572869389813415,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.2276,
|
|
"mean_token_accuracy": 0.03601204827427864,
|
|
"num_tokens": 182836.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.45373363494873,
|
|
"epoch": 0.0907715582450832,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1377,
|
|
"mean_token_accuracy": 0.02935194130986929,
|
|
"num_tokens": 194781.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.433523464202882,
|
|
"epoch": 0.09581442259203228,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9882,
|
|
"mean_token_accuracy": 0.03305415883660316,
|
|
"num_tokens": 206722.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.359123516082764,
|
|
"epoch": 0.10085728693898134,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8489,
|
|
"mean_token_accuracy": 0.04083778336644173,
|
|
"num_tokens": 217167.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.264568710327149,
|
|
"epoch": 0.1059001512859304,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.7978,
|
|
"mean_token_accuracy": 0.04325296822935343,
|
|
"num_tokens": 228427.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.185066509246827,
|
|
"epoch": 0.11094301563287948,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.6675,
|
|
"mean_token_accuracy": 0.04498021528124809,
|
|
"num_tokens": 239169.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.122811222076416,
|
|
"epoch": 0.11598587997982854,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.5549,
|
|
"mean_token_accuracy": 0.039299645088613035,
|
|
"num_tokens": 250393.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.999498176574708,
|
|
"epoch": 0.12102874432677761,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.3439,
|
|
"mean_token_accuracy": 0.049582514539361,
|
|
"num_tokens": 260573.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.756069564819336,
|
|
"epoch": 0.1260716086737267,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.2337,
|
|
"mean_token_accuracy": 0.0456912275403738,
|
|
"num_tokens": 271537.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.617351722717284,
|
|
"epoch": 0.13111447302067575,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.1765,
|
|
"mean_token_accuracy": 0.04553677849471569,
|
|
"num_tokens": 282080.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.397808265686034,
|
|
"epoch": 0.1361573373676248,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.0293,
|
|
"mean_token_accuracy": 0.04926962368190289,
|
|
"num_tokens": 292509.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.158194255828857,
|
|
"epoch": 0.14120020171457387,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9266,
|
|
"mean_token_accuracy": 0.04916851259768009,
|
|
"num_tokens": 303156.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.969250679016113,
|
|
"epoch": 0.14624306606152296,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.8665,
|
|
"mean_token_accuracy": 0.04366149418056011,
|
|
"num_tokens": 314398.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.68967514038086,
|
|
"epoch": 0.15128593040847202,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.894,
|
|
"mean_token_accuracy": 0.04541506320238113,
|
|
"num_tokens": 324774.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.563091468811034,
|
|
"epoch": 0.15632879475542108,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7359,
|
|
"mean_token_accuracy": 0.04416811689734459,
|
|
"num_tokens": 335154.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.345014476776123,
|
|
"epoch": 0.16137165910237014,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8209,
|
|
"mean_token_accuracy": 0.05068969242274761,
|
|
"num_tokens": 346180.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.28213872909546,
|
|
"epoch": 0.1664145234493192,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7329,
|
|
"mean_token_accuracy": 0.04978744387626648,
|
|
"num_tokens": 356722.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.216850662231446,
|
|
"epoch": 0.1714573877962683,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6846,
|
|
"mean_token_accuracy": 0.05312272384762764,
|
|
"num_tokens": 367245.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.235016822814941,
|
|
"epoch": 0.17650025214321735,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6401,
|
|
"mean_token_accuracy": 0.05696842148900032,
|
|
"num_tokens": 377434.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.149722671508789,
|
|
"epoch": 0.1815431164901664,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.5829,
|
|
"mean_token_accuracy": 0.06260009892284871,
|
|
"num_tokens": 388685.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.027026557922364,
|
|
"epoch": 0.18658598083711547,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5612,
|
|
"mean_token_accuracy": 0.05986677259206772,
|
|
"num_tokens": 400559.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.110166931152344,
|
|
"epoch": 0.19162884518406456,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.6123,
|
|
"mean_token_accuracy": 0.06047765538096428,
|
|
"num_tokens": 413059.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.079530906677245,
|
|
"epoch": 0.19667170953101362,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.5223,
|
|
"mean_token_accuracy": 0.06776017844676971,
|
|
"num_tokens": 424073.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.999411535263062,
|
|
"epoch": 0.20171457387796268,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4703,
|
|
"mean_token_accuracy": 0.06763602308928966,
|
|
"num_tokens": 435960.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.994884729385376,
|
|
"epoch": 0.20675743822491174,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4259,
|
|
"mean_token_accuracy": 0.06415937095880508,
|
|
"num_tokens": 445458.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.940965747833252,
|
|
"epoch": 0.2118003025718608,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.4427,
|
|
"mean_token_accuracy": 0.06450643800199032,
|
|
"num_tokens": 456237.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.971080732345581,
|
|
"epoch": 0.2168431669188099,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4559,
|
|
"mean_token_accuracy": 0.06802051141858101,
|
|
"num_tokens": 467210.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.95701413154602,
|
|
"epoch": 0.22188603126575895,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4836,
|
|
"mean_token_accuracy": 0.06283098589628935,
|
|
"num_tokens": 478447.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.983369398117065,
|
|
"epoch": 0.22692889561270801,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.4325,
|
|
"mean_token_accuracy": 0.06579711437225341,
|
|
"num_tokens": 489108.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.921357250213623,
|
|
"epoch": 0.23197175995965708,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.4914,
|
|
"mean_token_accuracy": 0.06151033192873001,
|
|
"num_tokens": 500889.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.812901639938355,
|
|
"epoch": 0.23701462430660616,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.3437,
|
|
"mean_token_accuracy": 0.06879591457545757,
|
|
"num_tokens": 511038.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.84229211807251,
|
|
"epoch": 0.24205748865355523,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4227,
|
|
"mean_token_accuracy": 0.06682286001741886,
|
|
"num_tokens": 522801.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.874321031570434,
|
|
"epoch": 0.24710035300050429,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.4278,
|
|
"mean_token_accuracy": 0.06629796028137207,
|
|
"num_tokens": 534795.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.927615451812744,
|
|
"epoch": 0.2521432173474534,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4248,
|
|
"mean_token_accuracy": 0.06450821422040462,
|
|
"num_tokens": 546656.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.768723440170288,
|
|
"epoch": 0.25718608169440244,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.29,
|
|
"mean_token_accuracy": 0.07409121245145797,
|
|
"num_tokens": 556673.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.753383922576904,
|
|
"epoch": 0.2622289460413515,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3377,
|
|
"mean_token_accuracy": 0.06640650182962418,
|
|
"num_tokens": 567044.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.754617547988891,
|
|
"epoch": 0.26727181038830056,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2684,
|
|
"mean_token_accuracy": 0.07174375429749488,
|
|
"num_tokens": 577305.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.622322416305542,
|
|
"epoch": 0.2723146747352496,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3679,
|
|
"mean_token_accuracy": 0.07188783846795559,
|
|
"num_tokens": 589285.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.7097578048706055,
|
|
"epoch": 0.2773575390821987,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1943,
|
|
"mean_token_accuracy": 0.07407240830361843,
|
|
"num_tokens": 598609.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.717816877365112,
|
|
"epoch": 0.28240040342914774,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2602,
|
|
"mean_token_accuracy": 0.07445880733430385,
|
|
"num_tokens": 609872.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.6907618045806885,
|
|
"epoch": 0.2874432677760968,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.2711,
|
|
"mean_token_accuracy": 0.08134151697158813,
|
|
"num_tokens": 622583.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.654936122894287,
|
|
"epoch": 0.2924861321230459,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2412,
|
|
"mean_token_accuracy": 0.07294923588633537,
|
|
"num_tokens": 633592.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.543200588226318,
|
|
"epoch": 0.297528996469995,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.169,
|
|
"mean_token_accuracy": 0.07622305229306221,
|
|
"num_tokens": 644858.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.596815156936645,
|
|
"epoch": 0.30257186081694404,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.245,
|
|
"mean_token_accuracy": 0.07087541855871678,
|
|
"num_tokens": 656438.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.674440145492554,
|
|
"epoch": 0.3076147251638931,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2518,
|
|
"mean_token_accuracy": 0.07656564638018608,
|
|
"num_tokens": 667942.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.500104904174805,
|
|
"epoch": 0.31265758951084216,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1468,
|
|
"mean_token_accuracy": 0.07846823334693909,
|
|
"num_tokens": 678596.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.610447406768799,
|
|
"epoch": 0.3177004538577912,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.3024,
|
|
"mean_token_accuracy": 0.06737342216074467,
|
|
"num_tokens": 691014.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.612302160263061,
|
|
"epoch": 0.3227433182047403,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2079,
|
|
"mean_token_accuracy": 0.07754847705364228,
|
|
"num_tokens": 701849.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.523909950256348,
|
|
"epoch": 0.32778618255168934,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1103,
|
|
"mean_token_accuracy": 0.08044524192810058,
|
|
"num_tokens": 712652.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.591261148452759,
|
|
"epoch": 0.3328290468986384,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1597,
|
|
"mean_token_accuracy": 0.07344280071556568,
|
|
"num_tokens": 723704.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.501336288452149,
|
|
"epoch": 0.3378719112455875,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1081,
|
|
"mean_token_accuracy": 0.07664402537047862,
|
|
"num_tokens": 734715.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.462927675247192,
|
|
"epoch": 0.3429147755925366,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1461,
|
|
"mean_token_accuracy": 0.07232181765139103,
|
|
"num_tokens": 746772.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.542045450210571,
|
|
"epoch": 0.34795763993948564,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1028,
|
|
"mean_token_accuracy": 0.07609748989343643,
|
|
"num_tokens": 757356.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.513222789764404,
|
|
"epoch": 0.3530005042864347,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1177,
|
|
"mean_token_accuracy": 0.07893777638673782,
|
|
"num_tokens": 768269.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.457395029067993,
|
|
"epoch": 0.35804336863338376,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0621,
|
|
"mean_token_accuracy": 0.07398857697844505,
|
|
"num_tokens": 779553.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.4384712219238285,
|
|
"epoch": 0.3630862329803328,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1477,
|
|
"mean_token_accuracy": 0.07859819531440734,
|
|
"num_tokens": 791466.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.519556713104248,
|
|
"epoch": 0.3681290973272819,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0409,
|
|
"mean_token_accuracy": 0.07564883790910244,
|
|
"num_tokens": 802064.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.367311239242554,
|
|
"epoch": 0.37317196167423095,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0721,
|
|
"mean_token_accuracy": 0.08020395338535309,
|
|
"num_tokens": 813534.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.481927680969238,
|
|
"epoch": 0.37821482602118,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0958,
|
|
"mean_token_accuracy": 0.07429472133517265,
|
|
"num_tokens": 824783.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.396484994888306,
|
|
"epoch": 0.3832576903681291,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0813,
|
|
"mean_token_accuracy": 0.0780322715640068,
|
|
"num_tokens": 835019.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.43559775352478,
|
|
"epoch": 0.3883005547150782,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9778,
|
|
"mean_token_accuracy": 0.07786969803273677,
|
|
"num_tokens": 845954.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.349666690826416,
|
|
"epoch": 0.39334341906202724,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9968,
|
|
"mean_token_accuracy": 0.08358137160539628,
|
|
"num_tokens": 856393.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.377739429473877,
|
|
"epoch": 0.3983862834089763,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0058,
|
|
"mean_token_accuracy": 0.08096945993602275,
|
|
"num_tokens": 868079.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.420732831954956,
|
|
"epoch": 0.40342914775592537,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0597,
|
|
"mean_token_accuracy": 0.08519057407975197,
|
|
"num_tokens": 878995.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.359525680541992,
|
|
"epoch": 0.4084720121028744,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.9767,
|
|
"mean_token_accuracy": 0.08249035105109215,
|
|
"num_tokens": 889159.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.397777795791626,
|
|
"epoch": 0.4135148764498235,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0906,
|
|
"mean_token_accuracy": 0.07337038293480873,
|
|
"num_tokens": 900186.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.36064920425415,
|
|
"epoch": 0.41855774079677255,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0054,
|
|
"mean_token_accuracy": 0.0781717125326395,
|
|
"num_tokens": 910713.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.512122440338135,
|
|
"epoch": 0.4236006051437216,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0956,
|
|
"mean_token_accuracy": 0.07817071452736854,
|
|
"num_tokens": 922020.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.32919135093689,
|
|
"epoch": 0.4286434694906707,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.0302,
|
|
"mean_token_accuracy": 0.07741615772247315,
|
|
"num_tokens": 933033.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.3811193943023685,
|
|
"epoch": 0.4336863338376198,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0623,
|
|
"mean_token_accuracy": 0.07646705359220504,
|
|
"num_tokens": 944394.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.382322645187378,
|
|
"epoch": 0.43872919818456885,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.9714,
|
|
"mean_token_accuracy": 0.0816558338701725,
|
|
"num_tokens": 955337.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.19914608001709,
|
|
"epoch": 0.4437720625315179,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.9435,
|
|
"mean_token_accuracy": 0.08770947605371475,
|
|
"num_tokens": 967020.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.364274787902832,
|
|
"epoch": 0.44881492687846697,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9686,
|
|
"mean_token_accuracy": 0.08233175501227379,
|
|
"num_tokens": 978223.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.2527648448944095,
|
|
"epoch": 0.45385779122541603,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9751,
|
|
"mean_token_accuracy": 0.08211390711367131,
|
|
"num_tokens": 990239.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.344982099533081,
|
|
"epoch": 0.4589006555723651,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8979,
|
|
"mean_token_accuracy": 0.08162759393453597,
|
|
"num_tokens": 1001834.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.236715698242188,
|
|
"epoch": 0.46394351991931415,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8744,
|
|
"mean_token_accuracy": 0.09091370068490505,
|
|
"num_tokens": 1012321.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.2280114650726315,
|
|
"epoch": 0.4689863842662632,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9796,
|
|
"mean_token_accuracy": 0.08122261688113212,
|
|
"num_tokens": 1024500.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.270635604858398,
|
|
"epoch": 0.47402924861321233,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8926,
|
|
"mean_token_accuracy": 0.08805500157177448,
|
|
"num_tokens": 1034825.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.1545487403869625,
|
|
"epoch": 0.4790721129601614,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.8648,
|
|
"mean_token_accuracy": 0.08775893896818161,
|
|
"num_tokens": 1046174.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.252149486541748,
|
|
"epoch": 0.48411497730711045,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7807,
|
|
"mean_token_accuracy": 0.09338528215885163,
|
|
"num_tokens": 1055930.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.144695091247558,
|
|
"epoch": 0.4891578416540595,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8153,
|
|
"mean_token_accuracy": 0.09350402504205704,
|
|
"num_tokens": 1066433.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.173965978622436,
|
|
"epoch": 0.49420070600100857,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9306,
|
|
"mean_token_accuracy": 0.08193769156932831,
|
|
"num_tokens": 1077303.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.231033754348755,
|
|
"epoch": 0.49924357034795763,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8453,
|
|
"mean_token_accuracy": 0.08849839866161346,
|
|
"num_tokens": 1088053.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.107387685775757,
|
|
"epoch": 0.5042864346949067,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8499,
|
|
"mean_token_accuracy": 0.09236783608794212,
|
|
"num_tokens": 1098518.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5042864346949067,
|
|
"eval_entropy": 7.061949415667973,
|
|
"eval_loss": 6.937669277191162,
|
|
"eval_mean_token_accuracy": 0.09019645222915341,
|
|
"eval_num_tokens": 1098518.0,
|
|
"eval_runtime": 2.6772,
|
|
"eval_samples_per_second": 1172.119,
|
|
"eval_steps_per_second": 146.795,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.228630304336548,
|
|
"epoch": 0.5093292990418558,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8577,
|
|
"mean_token_accuracy": 0.08858487010002136,
|
|
"num_tokens": 1109768.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.081269025802612,
|
|
"epoch": 0.5143721633888049,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.8846,
|
|
"mean_token_accuracy": 0.09013336524367332,
|
|
"num_tokens": 1121969.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.162287855148316,
|
|
"epoch": 0.5194150277357539,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7391,
|
|
"mean_token_accuracy": 0.09084992706775666,
|
|
"num_tokens": 1132405.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.103027248382569,
|
|
"epoch": 0.524457892082703,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.07809185199439525,
|
|
"num_tokens": 1143705.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.171495628356934,
|
|
"epoch": 0.529500756429652,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.8513,
|
|
"mean_token_accuracy": 0.08691945150494576,
|
|
"num_tokens": 1154445.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.117181491851807,
|
|
"epoch": 0.5345436207766011,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.9407,
|
|
"mean_token_accuracy": 0.08625975549221039,
|
|
"num_tokens": 1166895.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.119032716751098,
|
|
"epoch": 0.5395864851235502,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8261,
|
|
"mean_token_accuracy": 0.09287582933902741,
|
|
"num_tokens": 1176744.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.1486015796661375,
|
|
"epoch": 0.5446293494704992,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8097,
|
|
"mean_token_accuracy": 0.08768841177225113,
|
|
"num_tokens": 1187024.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.054621648788452,
|
|
"epoch": 0.5496722138174484,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8001,
|
|
"mean_token_accuracy": 0.09048704728484154,
|
|
"num_tokens": 1198799.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.101481819152832,
|
|
"epoch": 0.5547150781643974,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7588,
|
|
"mean_token_accuracy": 0.09443475604057312,
|
|
"num_tokens": 1208585.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.084757947921753,
|
|
"epoch": 0.5597579425113465,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8007,
|
|
"mean_token_accuracy": 0.09204660803079605,
|
|
"num_tokens": 1218268.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.034044981002808,
|
|
"epoch": 0.5648008068582955,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8063,
|
|
"mean_token_accuracy": 0.09505132585763931,
|
|
"num_tokens": 1229106.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.110476064682007,
|
|
"epoch": 0.5698436712052446,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7378,
|
|
"mean_token_accuracy": 0.09246841743588448,
|
|
"num_tokens": 1239741.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.986316823959351,
|
|
"epoch": 0.5748865355521936,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7035,
|
|
"mean_token_accuracy": 0.08977311328053475,
|
|
"num_tokens": 1249658.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.991911506652832,
|
|
"epoch": 0.5799293998991427,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7475,
|
|
"mean_token_accuracy": 0.09191381558775902,
|
|
"num_tokens": 1259731.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.12698917388916,
|
|
"epoch": 0.5849722642460918,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7476,
|
|
"mean_token_accuracy": 0.09400721192359925,
|
|
"num_tokens": 1270087.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.9526652812957765,
|
|
"epoch": 0.5900151285930408,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7166,
|
|
"mean_token_accuracy": 0.09413416013121605,
|
|
"num_tokens": 1282567.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.915704536437988,
|
|
"epoch": 0.59505799293999,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7789,
|
|
"mean_token_accuracy": 0.09547073394060135,
|
|
"num_tokens": 1294133.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.990864229202271,
|
|
"epoch": 0.600100857286939,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6449,
|
|
"mean_token_accuracy": 0.0973285436630249,
|
|
"num_tokens": 1305306.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.955234050750732,
|
|
"epoch": 0.6051437216338881,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7173,
|
|
"mean_token_accuracy": 0.09564844146370888,
|
|
"num_tokens": 1316379.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.972650957107544,
|
|
"epoch": 0.6101865859808371,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6907,
|
|
"mean_token_accuracy": 0.09237802252173424,
|
|
"num_tokens": 1328121.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.9388415813446045,
|
|
"epoch": 0.6152294503277862,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7319,
|
|
"mean_token_accuracy": 0.08763695284724235,
|
|
"num_tokens": 1338728.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.982970094680786,
|
|
"epoch": 0.6202723146747352,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.711,
|
|
"mean_token_accuracy": 0.0885362021625042,
|
|
"num_tokens": 1349126.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.916971921920776,
|
|
"epoch": 0.6253151790216843,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6068,
|
|
"mean_token_accuracy": 0.09231596812605858,
|
|
"num_tokens": 1359813.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.890164804458618,
|
|
"epoch": 0.6303580433686334,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7633,
|
|
"mean_token_accuracy": 0.09019490703940392,
|
|
"num_tokens": 1371872.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.952266883850098,
|
|
"epoch": 0.6354009077155824,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6359,
|
|
"mean_token_accuracy": 0.08751971200108528,
|
|
"num_tokens": 1382789.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.918084335327149,
|
|
"epoch": 0.6404437720625316,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7236,
|
|
"mean_token_accuracy": 0.09168680310249329,
|
|
"num_tokens": 1393419.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.888974285125732,
|
|
"epoch": 0.6454866364094806,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5702,
|
|
"mean_token_accuracy": 0.09415488988161087,
|
|
"num_tokens": 1404674.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.903821563720703,
|
|
"epoch": 0.6505295007564297,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7776,
|
|
"mean_token_accuracy": 0.0927918791770935,
|
|
"num_tokens": 1416252.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.834169626235962,
|
|
"epoch": 0.6555723651033787,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.584,
|
|
"mean_token_accuracy": 0.09287177100777626,
|
|
"num_tokens": 1426681.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.861099433898926,
|
|
"epoch": 0.6606152294503278,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.58,
|
|
"mean_token_accuracy": 0.09400497153401374,
|
|
"num_tokens": 1436640.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.8637463569641115,
|
|
"epoch": 0.6656580937972768,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6155,
|
|
"mean_token_accuracy": 0.09446947798132896,
|
|
"num_tokens": 1447183.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.829385280609131,
|
|
"epoch": 0.6707009581442259,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.6385,
|
|
"mean_token_accuracy": 0.0950157254934311,
|
|
"num_tokens": 1458351.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.804144191741943,
|
|
"epoch": 0.675743822491175,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6369,
|
|
"mean_token_accuracy": 0.09405813217163086,
|
|
"num_tokens": 1469522.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.76968903541565,
|
|
"epoch": 0.680786686838124,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.565,
|
|
"mean_token_accuracy": 0.09773774519562721,
|
|
"num_tokens": 1479947.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.893829965591431,
|
|
"epoch": 0.6858295511850732,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5879,
|
|
"mean_token_accuracy": 0.0951867401599884,
|
|
"num_tokens": 1490374.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.818552112579345,
|
|
"epoch": 0.6908724155320222,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.58,
|
|
"mean_token_accuracy": 0.10172139406204224,
|
|
"num_tokens": 1500939.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.806358861923218,
|
|
"epoch": 0.6959152798789713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6962,
|
|
"mean_token_accuracy": 0.09438932090997695,
|
|
"num_tokens": 1512803.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.681111001968384,
|
|
"epoch": 0.7009581442259203,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5066,
|
|
"mean_token_accuracy": 0.0982896327972412,
|
|
"num_tokens": 1523679.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.747470474243164,
|
|
"epoch": 0.7060010085728694,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.527,
|
|
"mean_token_accuracy": 0.09784084260463714,
|
|
"num_tokens": 1533892.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.73495717048645,
|
|
"epoch": 0.7110438729198184,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5033,
|
|
"mean_token_accuracy": 0.10459957495331765,
|
|
"num_tokens": 1543935.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.718096542358398,
|
|
"epoch": 0.7160867372667675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5985,
|
|
"mean_token_accuracy": 0.09408540651202202,
|
|
"num_tokens": 1555989.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.870676755905151,
|
|
"epoch": 0.7211296016137166,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5575,
|
|
"mean_token_accuracy": 0.09637962281703949,
|
|
"num_tokens": 1566355.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.679803085327149,
|
|
"epoch": 0.7261724659606656,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5917,
|
|
"mean_token_accuracy": 0.09732575789093971,
|
|
"num_tokens": 1578403.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.768120145797729,
|
|
"epoch": 0.7312153303076148,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6428,
|
|
"mean_token_accuracy": 0.09719825014472008,
|
|
"num_tokens": 1590072.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.76740460395813,
|
|
"epoch": 0.7362581946545638,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.5704,
|
|
"mean_token_accuracy": 0.10138242095708846,
|
|
"num_tokens": 1602051.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.809198188781738,
|
|
"epoch": 0.7413010590015129,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6497,
|
|
"mean_token_accuracy": 0.10376588925719261,
|
|
"num_tokens": 1615741.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.637604141235352,
|
|
"epoch": 0.7463439233484619,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.4981,
|
|
"mean_token_accuracy": 0.10018217116594315,
|
|
"num_tokens": 1627042.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.76273832321167,
|
|
"epoch": 0.751386787695411,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.564,
|
|
"mean_token_accuracy": 0.10186072885990143,
|
|
"num_tokens": 1638777.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.612446403503418,
|
|
"epoch": 0.75642965204236,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4761,
|
|
"mean_token_accuracy": 0.10184447392821312,
|
|
"num_tokens": 1650266.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.727019262313843,
|
|
"epoch": 0.7614725163893091,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5096,
|
|
"mean_token_accuracy": 0.10364911928772927,
|
|
"num_tokens": 1660500.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.626423454284668,
|
|
"epoch": 0.7665153807362582,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.10132494196295738,
|
|
"num_tokens": 1671685.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.698447322845459,
|
|
"epoch": 0.7715582450832073,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5676,
|
|
"mean_token_accuracy": 0.0920490249991417,
|
|
"num_tokens": 1681954.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.720503091812134,
|
|
"epoch": 0.7766011094301564,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5492,
|
|
"mean_token_accuracy": 0.10080667063593865,
|
|
"num_tokens": 1691761.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.661979913711548,
|
|
"epoch": 0.7816439737771054,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4826,
|
|
"mean_token_accuracy": 0.10153618305921555,
|
|
"num_tokens": 1701715.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.6159828186035154,
|
|
"epoch": 0.7866868381240545,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4372,
|
|
"mean_token_accuracy": 0.1015503615140915,
|
|
"num_tokens": 1714530.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.619305753707886,
|
|
"epoch": 0.7917297024710035,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.527,
|
|
"mean_token_accuracy": 0.10115180537104607,
|
|
"num_tokens": 1725766.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.576332902908325,
|
|
"epoch": 0.7967725668179526,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4528,
|
|
"mean_token_accuracy": 0.1076912261545658,
|
|
"num_tokens": 1738307.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.6750335693359375,
|
|
"epoch": 0.8018154311649016,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5231,
|
|
"mean_token_accuracy": 0.09890256971120834,
|
|
"num_tokens": 1749129.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.730860948562622,
|
|
"epoch": 0.8068582955118507,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5414,
|
|
"mean_token_accuracy": 0.0941914051771164,
|
|
"num_tokens": 1759426.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.552480602264405,
|
|
"epoch": 0.8119011598587998,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3638,
|
|
"mean_token_accuracy": 0.10434163883328437,
|
|
"num_tokens": 1770906.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.582395315170288,
|
|
"epoch": 0.8169440242057489,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5188,
|
|
"mean_token_accuracy": 0.10247278958559036,
|
|
"num_tokens": 1782424.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.616803169250488,
|
|
"epoch": 0.821986888552698,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5031,
|
|
"mean_token_accuracy": 0.1025144636631012,
|
|
"num_tokens": 1794331.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.614215946197509,
|
|
"epoch": 0.827029752899647,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4325,
|
|
"mean_token_accuracy": 0.09733226075768471,
|
|
"num_tokens": 1805395.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.563755941390991,
|
|
"epoch": 0.8320726172465961,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4082,
|
|
"mean_token_accuracy": 0.0999007374048233,
|
|
"num_tokens": 1815820.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.52943525314331,
|
|
"epoch": 0.8371154815935451,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.10219496190547943,
|
|
"num_tokens": 1825754.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.578064441680908,
|
|
"epoch": 0.8421583459404942,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4729,
|
|
"mean_token_accuracy": 0.10351292341947556,
|
|
"num_tokens": 1837899.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.625820207595825,
|
|
"epoch": 0.8472012102874432,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4095,
|
|
"mean_token_accuracy": 0.10529221072793007,
|
|
"num_tokens": 1847659.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.479169607162476,
|
|
"epoch": 0.8522440746343923,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3946,
|
|
"mean_token_accuracy": 0.1047472171485424,
|
|
"num_tokens": 1860126.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.462310934066773,
|
|
"epoch": 0.8572869389813415,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3943,
|
|
"mean_token_accuracy": 0.10278471186757088,
|
|
"num_tokens": 1870567.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.556047487258911,
|
|
"epoch": 0.8623298033282905,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5126,
|
|
"mean_token_accuracy": 0.09607306495308876,
|
|
"num_tokens": 1880274.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.602808713912964,
|
|
"epoch": 0.8673726676752396,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5002,
|
|
"mean_token_accuracy": 0.09916604906320572,
|
|
"num_tokens": 1891122.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.556145715713501,
|
|
"epoch": 0.8724155320221886,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4984,
|
|
"mean_token_accuracy": 0.09922838732600212,
|
|
"num_tokens": 1902629.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.619171953201294,
|
|
"epoch": 0.8774583963691377,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4293,
|
|
"mean_token_accuracy": 0.10180645212531089,
|
|
"num_tokens": 1913808.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.459190368652344,
|
|
"epoch": 0.8825012607160867,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4308,
|
|
"mean_token_accuracy": 0.10478962510824204,
|
|
"num_tokens": 1925973.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.569327783584595,
|
|
"epoch": 0.8875441250630358,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.11385421827435493,
|
|
"num_tokens": 1937808.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.516397142410279,
|
|
"epoch": 0.8925869894099848,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4027,
|
|
"mean_token_accuracy": 0.10975287705659867,
|
|
"num_tokens": 1948581.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.4545928001403805,
|
|
"epoch": 0.8976298537569339,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4461,
|
|
"mean_token_accuracy": 0.10252013728022576,
|
|
"num_tokens": 1959977.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.630584287643432,
|
|
"epoch": 0.902672718103883,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4874,
|
|
"mean_token_accuracy": 0.09852444604039193,
|
|
"num_tokens": 1971328.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.492565059661866,
|
|
"epoch": 0.9077155824508321,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.581,
|
|
"mean_token_accuracy": 0.10265940502285957,
|
|
"num_tokens": 1983805.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.608131027221679,
|
|
"epoch": 0.9127584467977812,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5166,
|
|
"mean_token_accuracy": 0.10072435140609741,
|
|
"num_tokens": 1994773.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.5182877540588375,
|
|
"epoch": 0.9178013111447302,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3935,
|
|
"mean_token_accuracy": 0.10560430958867073,
|
|
"num_tokens": 2005600.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.5030282497406,
|
|
"epoch": 0.9228441754916793,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3335,
|
|
"mean_token_accuracy": 0.10612398311495781,
|
|
"num_tokens": 2016580.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.493871545791626,
|
|
"epoch": 0.9278870398386283,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4296,
|
|
"mean_token_accuracy": 0.10680199563503265,
|
|
"num_tokens": 2030318.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.425577878952026,
|
|
"epoch": 0.9329299041855774,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.10729326754808426,
|
|
"num_tokens": 2041744.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.464505004882812,
|
|
"epoch": 0.9379727685325264,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4385,
|
|
"mean_token_accuracy": 0.10069913193583488,
|
|
"num_tokens": 2053628.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.5448229789733885,
|
|
"epoch": 0.9430156328794755,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.10871829986572265,
|
|
"num_tokens": 2065069.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.539431858062744,
|
|
"epoch": 0.9480584972264247,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.439,
|
|
"mean_token_accuracy": 0.10412655398249626,
|
|
"num_tokens": 2076792.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.4169426441192625,
|
|
"epoch": 0.9531013615733737,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.2986,
|
|
"mean_token_accuracy": 0.11213012337684632,
|
|
"num_tokens": 2087373.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.446739196777344,
|
|
"epoch": 0.9581442259203228,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3437,
|
|
"mean_token_accuracy": 0.10855904519557953,
|
|
"num_tokens": 2100846.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.357406377792358,
|
|
"epoch": 0.9631870902672718,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3536,
|
|
"mean_token_accuracy": 0.11016998365521431,
|
|
"num_tokens": 2110668.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.449956846237183,
|
|
"epoch": 0.9682299546142209,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.3388,
|
|
"mean_token_accuracy": 0.10505538731813431,
|
|
"num_tokens": 2120917.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.381419658660889,
|
|
"epoch": 0.9732728189611699,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3337,
|
|
"mean_token_accuracy": 0.10150417611002922,
|
|
"num_tokens": 2132929.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.349277019500732,
|
|
"epoch": 0.978315683308119,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.2908,
|
|
"mean_token_accuracy": 0.11067101880908012,
|
|
"num_tokens": 2145251.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.365310096740723,
|
|
"epoch": 0.983358547655068,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3134,
|
|
"mean_token_accuracy": 0.11331562921404839,
|
|
"num_tokens": 2154702.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.436642789840699,
|
|
"epoch": 0.9884014120020171,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.3472,
|
|
"mean_token_accuracy": 0.1113656260073185,
|
|
"num_tokens": 2165924.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.408423995971679,
|
|
"epoch": 0.9934442763489663,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3166,
|
|
"mean_token_accuracy": 0.11193023920059204,
|
|
"num_tokens": 2176933.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.351378726959228,
|
|
"epoch": 0.9984871406959153,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.408,
|
|
"mean_token_accuracy": 0.10353770330548287,
|
|
"num_tokens": 2188812.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.459409925672743,
|
|
"epoch": 1.0030257186081695,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1274,
|
|
"mean_token_accuracy": 0.1104872499903043,
|
|
"num_tokens": 2199238.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.32691593170166,
|
|
"epoch": 1.0080685829551186,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.157,
|
|
"mean_token_accuracy": 0.11325521096587181,
|
|
"num_tokens": 2210530.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0080685829551186,
|
|
"eval_entropy": 6.25044517662689,
|
|
"eval_loss": 6.406233310699463,
|
|
"eval_mean_token_accuracy": 0.11158415295050951,
|
|
"eval_num_tokens": 2210530.0,
|
|
"eval_runtime": 2.6239,
|
|
"eval_samples_per_second": 1195.933,
|
|
"eval_steps_per_second": 149.777,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.298683118820191,
|
|
"epoch": 1.0131114473020675,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999997762223182,
|
|
"loss": 6.1697,
|
|
"mean_token_accuracy": 0.10936820954084396,
|
|
"num_tokens": 2221721.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.3320824146270756,
|
|
"epoch": 1.0181543116490166,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999988671262489,
|
|
"loss": 6.2022,
|
|
"mean_token_accuracy": 0.11588318571448326,
|
|
"num_tokens": 2233571.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.387577676773072,
|
|
"epoch": 1.0231971759959657,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999972587285105,
|
|
"loss": 6.2302,
|
|
"mean_token_accuracy": 0.11754814460873604,
|
|
"num_tokens": 2243419.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.327431154251099,
|
|
"epoch": 1.0282400403429148,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999949510341015,
|
|
"loss": 6.1845,
|
|
"mean_token_accuracy": 0.1209840402007103,
|
|
"num_tokens": 2254461.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.275969505310059,
|
|
"epoch": 1.033282904689864,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999919440501947,
|
|
"loss": 6.1625,
|
|
"mean_token_accuracy": 0.11422310993075371,
|
|
"num_tokens": 2266107.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.320016622543335,
|
|
"epoch": 1.0383257690368128,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999882377861355,
|
|
"loss": 6.1998,
|
|
"mean_token_accuracy": 0.11143601089715957,
|
|
"num_tokens": 2276084.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.3366124629974365,
|
|
"epoch": 1.043368633383762,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999838322534433,
|
|
"loss": 6.3104,
|
|
"mean_token_accuracy": 0.10776894018054009,
|
|
"num_tokens": 2287364.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.361055326461792,
|
|
"epoch": 1.048411497730711,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999787274658107,
|
|
"loss": 6.165,
|
|
"mean_token_accuracy": 0.11821243688464164,
|
|
"num_tokens": 2298313.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.301340389251709,
|
|
"epoch": 1.0534543620776602,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999729234391032,
|
|
"loss": 6.1697,
|
|
"mean_token_accuracy": 0.11393777951598168,
|
|
"num_tokens": 2310126.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.267262697219849,
|
|
"epoch": 1.058497226424609,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049996642019136,
|
|
"loss": 6.1057,
|
|
"mean_token_accuracy": 0.11893928498029709,
|
|
"num_tokens": 2320830.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.188592100143433,
|
|
"epoch": 1.0635400907715582,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999592177427933,
|
|
"loss": 6.1674,
|
|
"mean_token_accuracy": 0.11705956310033798,
|
|
"num_tokens": 2332027.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.273964786529541,
|
|
"epoch": 1.0685829551185073,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999513161157885,
|
|
"loss": 6.1881,
|
|
"mean_token_accuracy": 0.11757974997162819,
|
|
"num_tokens": 2342989.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.2867076873779295,
|
|
"epoch": 1.0736258194654564,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999427153349041,
|
|
"loss": 6.1449,
|
|
"mean_token_accuracy": 0.11460617780685425,
|
|
"num_tokens": 2353321.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.249440431594849,
|
|
"epoch": 1.0786686838124055,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999334154268714,
|
|
"loss": 6.1869,
|
|
"mean_token_accuracy": 0.11248090788722039,
|
|
"num_tokens": 2363910.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.243444728851318,
|
|
"epoch": 1.0837115481593544,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999234164205949,
|
|
"loss": 6.1837,
|
|
"mean_token_accuracy": 0.11587119996547698,
|
|
"num_tokens": 2374589.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.413973808288574,
|
|
"epoch": 1.0887544125063036,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999127183471517,
|
|
"loss": 6.2527,
|
|
"mean_token_accuracy": 0.11490818783640862,
|
|
"num_tokens": 2384684.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.217401218414307,
|
|
"epoch": 1.0937972768532527,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004999013212397915,
|
|
"loss": 6.2538,
|
|
"mean_token_accuracy": 0.1154061272740364,
|
|
"num_tokens": 2396649.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.351813507080078,
|
|
"epoch": 1.0988401412002018,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998892251339368,
|
|
"loss": 6.1245,
|
|
"mean_token_accuracy": 0.12135570496320724,
|
|
"num_tokens": 2407696.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.272814846038818,
|
|
"epoch": 1.1038830055471507,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998764300671827,
|
|
"loss": 6.2113,
|
|
"mean_token_accuracy": 0.11851643472909927,
|
|
"num_tokens": 2418333.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.2627276420593265,
|
|
"epoch": 1.1089258698940998,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.2299,
|
|
"mean_token_accuracy": 0.11417210474610329,
|
|
"num_tokens": 2431081.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.3452956199646,
|
|
"epoch": 1.113968734241049,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998487432122179,
|
|
"loss": 6.1657,
|
|
"mean_token_accuracy": 0.12272334843873978,
|
|
"num_tokens": 2441055.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.271652364730835,
|
|
"epoch": 1.119011598587998,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998338515100585,
|
|
"loss": 6.2095,
|
|
"mean_token_accuracy": 0.11820752769708634,
|
|
"num_tokens": 2451933.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.253003406524658,
|
|
"epoch": 1.1240544629349472,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998182610191022,
|
|
"loss": 6.2196,
|
|
"mean_token_accuracy": 0.11624273434281349,
|
|
"num_tokens": 2463062.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.2633141040802,
|
|
"epoch": 1.129097327281896,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998019717878047,
|
|
"loss": 6.2218,
|
|
"mean_token_accuracy": 0.11095208078622817,
|
|
"num_tokens": 2473650.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.33440580368042,
|
|
"epoch": 1.1341401916288452,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499784983866793,
|
|
"loss": 6.2787,
|
|
"mean_token_accuracy": 0.11275165677070617,
|
|
"num_tokens": 2484510.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.183946895599365,
|
|
"epoch": 1.1391830559757943,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004997672973088661,
|
|
"loss": 6.1521,
|
|
"mean_token_accuracy": 0.11300064027309417,
|
|
"num_tokens": 2495881.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.259727573394775,
|
|
"epoch": 1.1442259203227434,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004997489121689944,
|
|
"loss": 6.1078,
|
|
"mean_token_accuracy": 0.11746827661991119,
|
|
"num_tokens": 2506657.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.106284093856812,
|
|
"epoch": 1.1492687846696923,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997298285043189,
|
|
"loss": 6.121,
|
|
"mean_token_accuracy": 0.12150057107210159,
|
|
"num_tokens": 2518082.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.241512632369995,
|
|
"epoch": 1.1543116490166414,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997100463741527,
|
|
"loss": 6.0919,
|
|
"mean_token_accuracy": 0.12284635081887245,
|
|
"num_tokens": 2528722.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.2802904605865475,
|
|
"epoch": 1.1593545133635905,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996895658399785,
|
|
"loss": 6.0979,
|
|
"mean_token_accuracy": 0.1160362608730793,
|
|
"num_tokens": 2539008.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.154987907409668,
|
|
"epoch": 1.1643973777105396,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996683869654507,
|
|
"loss": 6.1121,
|
|
"mean_token_accuracy": 0.1161638431251049,
|
|
"num_tokens": 2549445.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.135028886795044,
|
|
"epoch": 1.1694402420574885,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996465098163934,
|
|
"loss": 6.1583,
|
|
"mean_token_accuracy": 0.11396335363388062,
|
|
"num_tokens": 2561079.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.272411060333252,
|
|
"epoch": 1.1744831064044376,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004996239344608015,
|
|
"loss": 6.0888,
|
|
"mean_token_accuracy": 0.12005198225378991,
|
|
"num_tokens": 2571540.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.181358337402344,
|
|
"epoch": 1.1795259707513868,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996006609688398,
|
|
"loss": 6.0589,
|
|
"mean_token_accuracy": 0.11994132474064827,
|
|
"num_tokens": 2582184.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.147908496856689,
|
|
"epoch": 1.1845688350983359,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995766894128426,
|
|
"loss": 6.1837,
|
|
"mean_token_accuracy": 0.12015488147735595,
|
|
"num_tokens": 2592771.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.2181120872497555,
|
|
"epoch": 1.189611699445285,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995520198673143,
|
|
"loss": 6.0737,
|
|
"mean_token_accuracy": 0.11993672326207161,
|
|
"num_tokens": 2602974.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.233900117874145,
|
|
"epoch": 1.194654563792234,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004995266524089283,
|
|
"loss": 6.06,
|
|
"mean_token_accuracy": 0.12310326918959617,
|
|
"num_tokens": 2613915.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.168328189849854,
|
|
"epoch": 1.199697428139183,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004995005871165271,
|
|
"loss": 6.1496,
|
|
"mean_token_accuracy": 0.1177591860294342,
|
|
"num_tokens": 2626038.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.194040536880493,
|
|
"epoch": 1.2047402924861321,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004994738240711225,
|
|
"loss": 6.1042,
|
|
"mean_token_accuracy": 0.12066976353526115,
|
|
"num_tokens": 2636448.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.251908779144287,
|
|
"epoch": 1.2097831568330812,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004994463633558945,
|
|
"loss": 6.1974,
|
|
"mean_token_accuracy": 0.11828442364931106,
|
|
"num_tokens": 2648803.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.1560954570770265,
|
|
"epoch": 1.2148260211800301,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994182050561917,
|
|
"loss": 6.0962,
|
|
"mean_token_accuracy": 0.1210496187210083,
|
|
"num_tokens": 2659981.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.110389518737793,
|
|
"epoch": 1.2198688855269793,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004993893492595306,
|
|
"loss": 6.0856,
|
|
"mean_token_accuracy": 0.12283521145582199,
|
|
"num_tokens": 2670340.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.237795495986939,
|
|
"epoch": 1.2249117498739284,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004993597960555959,
|
|
"loss": 6.1174,
|
|
"mean_token_accuracy": 0.12508749216794968,
|
|
"num_tokens": 2681147.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.172670507431031,
|
|
"epoch": 1.2299546142208775,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004993295455362395,
|
|
"loss": 6.0388,
|
|
"mean_token_accuracy": 0.1264190822839737,
|
|
"num_tokens": 2693877.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.102581548690796,
|
|
"epoch": 1.2349974785678266,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992985977954808,
|
|
"loss": 6.0757,
|
|
"mean_token_accuracy": 0.11901459470391273,
|
|
"num_tokens": 2704371.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.239574670791626,
|
|
"epoch": 1.2400403429147757,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000499266952929506,
|
|
"loss": 6.1372,
|
|
"mean_token_accuracy": 0.11878894343972206,
|
|
"num_tokens": 2714549.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.153753709793091,
|
|
"epoch": 1.2450832072617246,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992346110366682,
|
|
"loss": 6.0684,
|
|
"mean_token_accuracy": 0.12529382407665252,
|
|
"num_tokens": 2723750.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.140466403961182,
|
|
"epoch": 1.2501260716086737,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004992015722174865,
|
|
"loss": 6.1064,
|
|
"mean_token_accuracy": 0.11989799737930298,
|
|
"num_tokens": 2734557.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.191011524200439,
|
|
"epoch": 1.2551689359556228,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004991678365746466,
|
|
"loss": 6.1197,
|
|
"mean_token_accuracy": 0.12287942469120025,
|
|
"num_tokens": 2746521.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.118342781066895,
|
|
"epoch": 1.2602118003025717,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004991334042129996,
|
|
"loss": 6.1396,
|
|
"mean_token_accuracy": 0.12459613531827926,
|
|
"num_tokens": 2758074.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.138072919845581,
|
|
"epoch": 1.2652546646495209,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004990982752395618,
|
|
"loss": 5.9861,
|
|
"mean_token_accuracy": 0.12695645466446875,
|
|
"num_tokens": 2770059.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.090990257263184,
|
|
"epoch": 1.27029752899647,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004990624497635152,
|
|
"loss": 6.0559,
|
|
"mean_token_accuracy": 0.12386075705289841,
|
|
"num_tokens": 2780802.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.221565437316895,
|
|
"epoch": 1.275340393343419,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.1523,
|
|
"mean_token_accuracy": 0.12033093869686126,
|
|
"num_tokens": 2793693.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.183721590042114,
|
|
"epoch": 1.2803832576903682,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004989887097511451,
|
|
"loss": 6.1682,
|
|
"mean_token_accuracy": 0.11709739565849304,
|
|
"num_tokens": 2805547.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.118980169296265,
|
|
"epoch": 1.2854261220373173,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989507954440074,
|
|
"loss": 6.0879,
|
|
"mean_token_accuracy": 0.117892437428236,
|
|
"num_tokens": 2816207.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.165093946456909,
|
|
"epoch": 1.2904689863842662,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004989121850926313,
|
|
"loss": 6.0669,
|
|
"mean_token_accuracy": 0.11663686856627464,
|
|
"num_tokens": 2827272.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.1378819942474365,
|
|
"epoch": 1.2955118507312153,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004988728788170188,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.12149282097816468,
|
|
"num_tokens": 2839814.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.1398155212402346,
|
|
"epoch": 1.3005547150781644,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004988328767393346,
|
|
"loss": 6.1763,
|
|
"mean_token_accuracy": 0.11872457265853882,
|
|
"num_tokens": 2851629.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.199816131591797,
|
|
"epoch": 1.3055975794251133,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004987921789839061,
|
|
"loss": 6.1082,
|
|
"mean_token_accuracy": 0.12454739883542061,
|
|
"num_tokens": 2863518.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.043096208572388,
|
|
"epoch": 1.3106404437720625,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004987507856772229,
|
|
"loss": 6.0624,
|
|
"mean_token_accuracy": 0.1212753787636757,
|
|
"num_tokens": 2874803.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.212449645996093,
|
|
"epoch": 1.3156833081190116,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987086969479364,
|
|
"loss": 6.05,
|
|
"mean_token_accuracy": 0.12336436361074447,
|
|
"num_tokens": 2886052.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.058530807495117,
|
|
"epoch": 1.3207261724659607,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986659129268592,
|
|
"loss": 6.0717,
|
|
"mean_token_accuracy": 0.12124264538288117,
|
|
"num_tokens": 2897797.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.182499027252197,
|
|
"epoch": 1.3257690368129098,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004986224337469652,
|
|
"loss": 6.0392,
|
|
"mean_token_accuracy": 0.11823606640100479,
|
|
"num_tokens": 2908752.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.041653299331665,
|
|
"epoch": 1.330811901159859,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004985782595433886,
|
|
"loss": 6.1067,
|
|
"mean_token_accuracy": 0.11841807141900063,
|
|
"num_tokens": 2919351.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.149489307403565,
|
|
"epoch": 1.3358547655068078,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 6.0661,
|
|
"mean_token_accuracy": 0.12007037401199341,
|
|
"num_tokens": 2930198.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.100072622299194,
|
|
"epoch": 1.340897629853757,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000498487826616525,
|
|
"loss": 6.0666,
|
|
"mean_token_accuracy": 0.12396523952484131,
|
|
"num_tokens": 2941349.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.110146522521973,
|
|
"epoch": 1.345940494200706,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004984415681743056,
|
|
"loss": 5.9878,
|
|
"mean_token_accuracy": 0.12134144976735114,
|
|
"num_tokens": 2951740.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.1147205352783205,
|
|
"epoch": 1.350983358547655,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004983946152705382,
|
|
"loss": 6.0311,
|
|
"mean_token_accuracy": 0.12353800013661384,
|
|
"num_tokens": 2963900.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.1047674179077145,
|
|
"epoch": 1.356026222894604,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004983469680511531,
|
|
"loss": 6.0708,
|
|
"mean_token_accuracy": 0.11983567550778389,
|
|
"num_tokens": 2975385.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.116179466247559,
|
|
"epoch": 1.3610690872415532,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004982986266642393,
|
|
"loss": 6.0633,
|
|
"mean_token_accuracy": 0.12223733887076378,
|
|
"num_tokens": 2986340.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.080137825012207,
|
|
"epoch": 1.3661119515885023,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004982495912600426,
|
|
"loss": 5.9413,
|
|
"mean_token_accuracy": 0.128122878074646,
|
|
"num_tokens": 2996582.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.022544431686401,
|
|
"epoch": 1.3711548159354514,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004981998619909664,
|
|
"loss": 5.9682,
|
|
"mean_token_accuracy": 0.12417263835668564,
|
|
"num_tokens": 3007620.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.088971042633057,
|
|
"epoch": 1.3761976802824005,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004981494390115704,
|
|
"loss": 6.0683,
|
|
"mean_token_accuracy": 0.12460343763232232,
|
|
"num_tokens": 3019520.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.027232551574707,
|
|
"epoch": 1.3812405446293494,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004980983224785704,
|
|
"loss": 6.0585,
|
|
"mean_token_accuracy": 0.11659184172749519,
|
|
"num_tokens": 3030616.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.165927600860596,
|
|
"epoch": 1.3862834089762985,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004980465125508376,
|
|
"loss": 6.1292,
|
|
"mean_token_accuracy": 0.11850353851914405,
|
|
"num_tokens": 3042383.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.158199501037598,
|
|
"epoch": 1.3913262733232477,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004979940093893986,
|
|
"loss": 6.0277,
|
|
"mean_token_accuracy": 0.1222493290901184,
|
|
"num_tokens": 3052833.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.009306812286377,
|
|
"epoch": 1.3963691376701965,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004979408131574345,
|
|
"loss": 6.0017,
|
|
"mean_token_accuracy": 0.12259870544075965,
|
|
"num_tokens": 3063800.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.128568363189697,
|
|
"epoch": 1.4014120020171457,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004978869240202802,
|
|
"loss": 6.1616,
|
|
"mean_token_accuracy": 0.12267429605126381,
|
|
"num_tokens": 3075243.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.081125736236572,
|
|
"epoch": 1.4064548663640948,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004978323421454246,
|
|
"loss": 6.001,
|
|
"mean_token_accuracy": 0.1273365207016468,
|
|
"num_tokens": 3086590.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.090214252471924,
|
|
"epoch": 1.411497730711044,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004977770677025092,
|
|
"loss": 6.0373,
|
|
"mean_token_accuracy": 0.1295775778591633,
|
|
"num_tokens": 3097554.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.092472696304322,
|
|
"epoch": 1.416540595057993,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004977211008633284,
|
|
"loss": 6.057,
|
|
"mean_token_accuracy": 0.12078494280576706,
|
|
"num_tokens": 3110051.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.035629320144653,
|
|
"epoch": 1.4215834594049421,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004976644418018285,
|
|
"loss": 5.9504,
|
|
"mean_token_accuracy": 0.12921829596161843,
|
|
"num_tokens": 3120640.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.042394065856934,
|
|
"epoch": 1.426626323751891,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000497607090694107,
|
|
"loss": 5.9583,
|
|
"mean_token_accuracy": 0.12378115579485893,
|
|
"num_tokens": 3131293.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.968035554885864,
|
|
"epoch": 1.4316691880988401,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004975490477184126,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.12400781959295273,
|
|
"num_tokens": 3141272.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.159467649459839,
|
|
"epoch": 1.4367120524457893,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000497490313055144,
|
|
"loss": 5.9489,
|
|
"mean_token_accuracy": 0.12632449343800545,
|
|
"num_tokens": 3151349.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.933459615707397,
|
|
"epoch": 1.4417549167927382,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 6.0528,
|
|
"mean_token_accuracy": 0.12241479903459548,
|
|
"num_tokens": 3162258.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.177392387390137,
|
|
"epoch": 1.4467977811396873,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004973707693982287,
|
|
"loss": 6.0687,
|
|
"mean_token_accuracy": 0.12557331547141076,
|
|
"num_tokens": 3172821.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.031765031814575,
|
|
"epoch": 1.4518406454866364,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004973099607761264,
|
|
"loss": 5.9993,
|
|
"mean_token_accuracy": 0.12058438062667846,
|
|
"num_tokens": 3183720.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.232817029953003,
|
|
"epoch": 1.4568835098335855,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004972484612095378,
|
|
"loss": 6.0611,
|
|
"mean_token_accuracy": 0.12563046663999558,
|
|
"num_tokens": 3192882.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.00401120185852,
|
|
"epoch": 1.4619263741805346,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497186270889605,
|
|
"loss": 6.0029,
|
|
"mean_token_accuracy": 0.12171068787574768,
|
|
"num_tokens": 3203908.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.086980676651001,
|
|
"epoch": 1.4669692385274837,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004971233900096169,
|
|
"loss": 6.0922,
|
|
"mean_token_accuracy": 0.11998873427510262,
|
|
"num_tokens": 3215145.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.075078773498535,
|
|
"epoch": 1.4720121028744326,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004970598187650085,
|
|
"loss": 5.9639,
|
|
"mean_token_accuracy": 0.13046645373106003,
|
|
"num_tokens": 3225714.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.057655906677246,
|
|
"epoch": 1.4770549672213817,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000496995557353361,
|
|
"loss": 6.0777,
|
|
"mean_token_accuracy": 0.12587454319000244,
|
|
"num_tokens": 3238541.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.956527376174927,
|
|
"epoch": 1.4820978315683309,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004969306059744004,
|
|
"loss": 5.8903,
|
|
"mean_token_accuracy": 0.1326131209731102,
|
|
"num_tokens": 3250219.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.961444568634033,
|
|
"epoch": 1.4871406959152798,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004968649648299968,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.12518008425831795,
|
|
"num_tokens": 3261676.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.040499210357666,
|
|
"epoch": 1.4921835602622289,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004967986341241646,
|
|
"loss": 6.0356,
|
|
"mean_token_accuracy": 0.11670267432928086,
|
|
"num_tokens": 3272862.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.975739526748657,
|
|
"epoch": 1.497226424609178,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004967316140630611,
|
|
"loss": 5.9459,
|
|
"mean_token_accuracy": 0.12727636694908143,
|
|
"num_tokens": 3282891.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.996143436431884,
|
|
"epoch": 1.502269288956127,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004966639048549862,
|
|
"loss": 5.8887,
|
|
"mean_token_accuracy": 0.12851305529475213,
|
|
"num_tokens": 3293105.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.014690017700195,
|
|
"epoch": 1.5073121533030762,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004965955067103816,
|
|
"loss": 5.959,
|
|
"mean_token_accuracy": 0.13141607046127318,
|
|
"num_tokens": 3304431.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.129389524459839,
|
|
"epoch": 1.5123550176500253,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004965264198418305,
|
|
"loss": 6.0494,
|
|
"mean_token_accuracy": 0.12193582206964493,
|
|
"num_tokens": 3314391.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.5123550176500253,
|
|
"eval_entropy": 5.9836117936151325,
|
|
"eval_loss": 6.151969909667969,
|
|
"eval_mean_token_accuracy": 0.12492486130228177,
|
|
"eval_num_tokens": 3314391.0,
|
|
"eval_runtime": 2.6419,
|
|
"eval_samples_per_second": 1187.797,
|
|
"eval_steps_per_second": 148.759,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.002770566940308,
|
|
"epoch": 1.5173978819969742,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004964566444640564,
|
|
"loss": 5.9062,
|
|
"mean_token_accuracy": 0.1287309117615223,
|
|
"num_tokens": 3324578.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.993435716629028,
|
|
"epoch": 1.5224407463439233,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004963861807939227,
|
|
"loss": 5.9685,
|
|
"mean_token_accuracy": 0.13316882848739625,
|
|
"num_tokens": 3334724.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.05625205039978,
|
|
"epoch": 1.5274836106908725,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004963150290504321,
|
|
"loss": 6.0687,
|
|
"mean_token_accuracy": 0.1278290867805481,
|
|
"num_tokens": 3345371.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.040426445007324,
|
|
"epoch": 1.5325264750378214,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004962431894547259,
|
|
"loss": 5.9856,
|
|
"mean_token_accuracy": 0.12599704489111901,
|
|
"num_tokens": 3357176.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.967979764938354,
|
|
"epoch": 1.5375693393847705,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004961706622300834,
|
|
"loss": 6.0129,
|
|
"mean_token_accuracy": 0.11759765893220901,
|
|
"num_tokens": 3368213.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.0974396705627445,
|
|
"epoch": 1.5426122037317196,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004960974476019206,
|
|
"loss": 5.9772,
|
|
"mean_token_accuracy": 0.12633119896054268,
|
|
"num_tokens": 3379676.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.973466157913208,
|
|
"epoch": 1.5476550680786687,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004960235457977905,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.12467504814267158,
|
|
"num_tokens": 3390765.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.052669954299927,
|
|
"epoch": 1.5526979324256178,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004959489570473814,
|
|
"loss": 5.9436,
|
|
"mean_token_accuracy": 0.1279728040099144,
|
|
"num_tokens": 3401804.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.094581937789917,
|
|
"epoch": 1.557740796772567,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000495873681582517,
|
|
"loss": 6.0648,
|
|
"mean_token_accuracy": 0.1282322771847248,
|
|
"num_tokens": 3413509.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.933273983001709,
|
|
"epoch": 1.5627836611195158,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004957977196371552,
|
|
"loss": 5.9076,
|
|
"mean_token_accuracy": 0.13429276049137115,
|
|
"num_tokens": 3423807.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.946268749237061,
|
|
"epoch": 1.567826525466465,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004957210714473873,
|
|
"loss": 5.88,
|
|
"mean_token_accuracy": 0.12932645455002784,
|
|
"num_tokens": 3434214.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.094040393829346,
|
|
"epoch": 1.572869389813414,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004956437372514378,
|
|
"loss": 5.9502,
|
|
"mean_token_accuracy": 0.12896995097398758,
|
|
"num_tokens": 3445081.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.898585033416748,
|
|
"epoch": 1.577912254160363,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004955657172896631,
|
|
"loss": 5.9275,
|
|
"mean_token_accuracy": 0.13340998068451881,
|
|
"num_tokens": 3454992.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.992181396484375,
|
|
"epoch": 1.582955118507312,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004954870118045511,
|
|
"loss": 6.0237,
|
|
"mean_token_accuracy": 0.11989426463842393,
|
|
"num_tokens": 3465635.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.947319221496582,
|
|
"epoch": 1.5879979828542612,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004954076210407203,
|
|
"loss": 5.9283,
|
|
"mean_token_accuracy": 0.13070602416992189,
|
|
"num_tokens": 3477746.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.0351625919342045,
|
|
"epoch": 1.5930408472012103,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004953275452449188,
|
|
"loss": 5.9775,
|
|
"mean_token_accuracy": 0.12918647080659867,
|
|
"num_tokens": 3488915.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.954270458221435,
|
|
"epoch": 1.5980837115481594,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004952467846660243,
|
|
"loss": 5.9517,
|
|
"mean_token_accuracy": 0.12838053107261657,
|
|
"num_tokens": 3499096.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.958280849456787,
|
|
"epoch": 1.6031265758951085,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004951653395550423,
|
|
"loss": 5.8424,
|
|
"mean_token_accuracy": 0.12435204163193703,
|
|
"num_tokens": 3512544.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.980319499969482,
|
|
"epoch": 1.6081694402420574,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.9792,
|
|
"mean_token_accuracy": 0.1283943310379982,
|
|
"num_tokens": 3522814.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.901864099502563,
|
|
"epoch": 1.6132123045890066,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004950003967514762,
|
|
"loss": 5.9033,
|
|
"mean_token_accuracy": 0.13048650473356246,
|
|
"num_tokens": 3533724.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.041346263885498,
|
|
"epoch": 1.6182551689359557,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000494916899571538,
|
|
"loss": 5.9635,
|
|
"mean_token_accuracy": 0.1256171390414238,
|
|
"num_tokens": 3544777.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.030082273483276,
|
|
"epoch": 1.6232980332829046,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004948327188848029,
|
|
"loss": 5.9863,
|
|
"mean_token_accuracy": 0.1326717607676983,
|
|
"num_tokens": 3556397.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.949783515930176,
|
|
"epoch": 1.6283408976298537,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004947478549529065,
|
|
"loss": 5.9397,
|
|
"mean_token_accuracy": 0.12490741759538651,
|
|
"num_tokens": 3568320.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.998614501953125,
|
|
"epoch": 1.6333837619768028,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004946623080396077,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.12845998406410217,
|
|
"num_tokens": 3578289.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.963990306854248,
|
|
"epoch": 1.638426626323752,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004945760784107884,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.12893664538860322,
|
|
"num_tokens": 3589710.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.021360588073731,
|
|
"epoch": 1.643469490670701,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004944891663344522,
|
|
"loss": 5.9729,
|
|
"mean_token_accuracy": 0.12624122723937034,
|
|
"num_tokens": 3601618.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.045843553543091,
|
|
"epoch": 1.6485123550176501,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004944015720807239,
|
|
"loss": 6.0503,
|
|
"mean_token_accuracy": 0.1247016966342926,
|
|
"num_tokens": 3612329.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.013518476486206,
|
|
"epoch": 1.653555219364599,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004943132959218484,
|
|
"loss": 6.0217,
|
|
"mean_token_accuracy": 0.12285120636224747,
|
|
"num_tokens": 3623887.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.96614294052124,
|
|
"epoch": 1.6585980837115482,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00049422433813219,
|
|
"loss": 5.9486,
|
|
"mean_token_accuracy": 0.12361961752176284,
|
|
"num_tokens": 3635484.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.967590427398681,
|
|
"epoch": 1.6636409480584973,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004941346989882318,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.12881785854697228,
|
|
"num_tokens": 3645802.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.977982473373413,
|
|
"epoch": 1.6686838124054462,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004940443787685739,
|
|
"loss": 5.9296,
|
|
"mean_token_accuracy": 0.12940604165196418,
|
|
"num_tokens": 3655599.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.026707506179809,
|
|
"epoch": 1.6737266767523953,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004939533777539341,
|
|
"loss": 5.9475,
|
|
"mean_token_accuracy": 0.12480253130197524,
|
|
"num_tokens": 3667019.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.816624021530151,
|
|
"epoch": 1.6787695410993444,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004938616962271453,
|
|
"loss": 5.8174,
|
|
"mean_token_accuracy": 0.13443900346755983,
|
|
"num_tokens": 3678004.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.990376710891724,
|
|
"epoch": 1.6838124054462935,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004937693344731561,
|
|
"loss": 5.8891,
|
|
"mean_token_accuracy": 0.1256047248840332,
|
|
"num_tokens": 3688177.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.9272315979003904,
|
|
"epoch": 1.6888552697932426,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004936762927790286,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.1307574525475502,
|
|
"num_tokens": 3700375.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.980884552001953,
|
|
"epoch": 1.6938981341401917,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004935825714339389,
|
|
"loss": 5.9813,
|
|
"mean_token_accuracy": 0.1269807793200016,
|
|
"num_tokens": 3711629.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.915404605865478,
|
|
"epoch": 1.6989409984871406,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004934881707291748,
|
|
"loss": 5.8829,
|
|
"mean_token_accuracy": 0.12866655737161636,
|
|
"num_tokens": 3723570.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.964293813705444,
|
|
"epoch": 1.7039838628340898,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004933930909581362,
|
|
"loss": 5.8167,
|
|
"mean_token_accuracy": 0.13202881217002868,
|
|
"num_tokens": 3733687.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.902900743484497,
|
|
"epoch": 1.7090267271810389,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004932973324163328,
|
|
"loss": 5.8501,
|
|
"mean_token_accuracy": 0.13000352382659913,
|
|
"num_tokens": 3744455.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.912310600280762,
|
|
"epoch": 1.7140695915279878,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004932008954013846,
|
|
"loss": 5.8379,
|
|
"mean_token_accuracy": 0.12738796770572663,
|
|
"num_tokens": 3756079.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.903970384597779,
|
|
"epoch": 1.7191124558749369,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004931037802130201,
|
|
"loss": 5.8167,
|
|
"mean_token_accuracy": 0.13375495001673698,
|
|
"num_tokens": 3766548.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.9737762928009035,
|
|
"epoch": 1.724155320221886,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004930059871530755,
|
|
"loss": 5.9092,
|
|
"mean_token_accuracy": 0.13129813969135284,
|
|
"num_tokens": 3776482.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.92631573677063,
|
|
"epoch": 1.7291981845688351,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004929075165254938,
|
|
"loss": 5.8892,
|
|
"mean_token_accuracy": 0.1338859997689724,
|
|
"num_tokens": 3787030.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.911603832244873,
|
|
"epoch": 1.7342410489157842,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004928083686363241,
|
|
"loss": 5.8502,
|
|
"mean_token_accuracy": 0.12917665019631386,
|
|
"num_tokens": 3797575.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.90240626335144,
|
|
"epoch": 1.7392839132627333,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004927085437937204,
|
|
"loss": 5.8342,
|
|
"mean_token_accuracy": 0.13536562025547028,
|
|
"num_tokens": 3807634.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.942410898208618,
|
|
"epoch": 1.7443267776096822,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 6.0215,
|
|
"mean_token_accuracy": 0.12834933772683144,
|
|
"num_tokens": 3820691.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.9908922672271725,
|
|
"epoch": 1.7493696419566314,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000492506864491345,
|
|
"loss": 6.0174,
|
|
"mean_token_accuracy": 0.12165129706263542,
|
|
"num_tokens": 3832434.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.99032940864563,
|
|
"epoch": 1.7544125063035805,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004924050106583976,
|
|
"loss": 5.9201,
|
|
"mean_token_accuracy": 0.13164895102381707,
|
|
"num_tokens": 3842547.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.9721756935119625,
|
|
"epoch": 1.7594553706505294,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004923024811256618,
|
|
"loss": 5.9257,
|
|
"mean_token_accuracy": 0.12628312036395073,
|
|
"num_tokens": 3852830.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.966777801513672,
|
|
"epoch": 1.7644982349974785,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492199276211802,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.12151766121387482,
|
|
"num_tokens": 3863396.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.879304456710815,
|
|
"epoch": 1.7695410993444276,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004920953962375813,
|
|
"loss": 5.8456,
|
|
"mean_token_accuracy": 0.12973029911518097,
|
|
"num_tokens": 3874380.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.922623920440674,
|
|
"epoch": 1.7745839636913767,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 5.8587,
|
|
"mean_token_accuracy": 0.1308340273797512,
|
|
"num_tokens": 3883454.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.874415349960327,
|
|
"epoch": 1.7796268280383258,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004918856124016002,
|
|
"loss": 5.8596,
|
|
"mean_token_accuracy": 0.1265338569879532,
|
|
"num_tokens": 3893419.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.949326276779175,
|
|
"epoch": 1.784669692385275,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004917797091918526,
|
|
"loss": 5.9122,
|
|
"mean_token_accuracy": 0.12785057127475738,
|
|
"num_tokens": 3903760.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.9434820175170895,
|
|
"epoch": 1.789712556732224,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004916731322257683,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.13075582906603814,
|
|
"num_tokens": 3914881.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.076316213607788,
|
|
"epoch": 1.794755421079173,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004915658818345908,
|
|
"loss": 5.9866,
|
|
"mean_token_accuracy": 0.12632244527339936,
|
|
"num_tokens": 3925774.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.857404661178589,
|
|
"epoch": 1.799798285426122,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004914579583516568,
|
|
"loss": 5.8635,
|
|
"mean_token_accuracy": 0.13342424854636192,
|
|
"num_tokens": 3936357.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.05221209526062,
|
|
"epoch": 1.804841149773071,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000491349362112395,
|
|
"loss": 5.9406,
|
|
"mean_token_accuracy": 0.130137200653553,
|
|
"num_tokens": 3946493.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.943054580688477,
|
|
"epoch": 1.80988401412002,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000491240093454325,
|
|
"loss": 5.918,
|
|
"mean_token_accuracy": 0.12943230643868447,
|
|
"num_tokens": 3958022.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.880387878417968,
|
|
"epoch": 1.8149268784669692,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004911301527170563,
|
|
"loss": 5.74,
|
|
"mean_token_accuracy": 0.13886643648147584,
|
|
"num_tokens": 3968033.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.889808988571167,
|
|
"epoch": 1.8199697428139183,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004910195402422873,
|
|
"loss": 5.9123,
|
|
"mean_token_accuracy": 0.1269322395324707,
|
|
"num_tokens": 3978319.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.946580696105957,
|
|
"epoch": 1.8250126071608674,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004909082563738042,
|
|
"loss": 5.8572,
|
|
"mean_token_accuracy": 0.1318647786974907,
|
|
"num_tokens": 3989519.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.929449892044067,
|
|
"epoch": 1.8300554715078166,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004907963014574797,
|
|
"loss": 5.943,
|
|
"mean_token_accuracy": 0.13093527406454086,
|
|
"num_tokens": 4000680.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.897009038925171,
|
|
"epoch": 1.8350983358547657,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004906836758412722,
|
|
"loss": 5.8782,
|
|
"mean_token_accuracy": 0.13037897050380706,
|
|
"num_tokens": 4012886.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.903923892974854,
|
|
"epoch": 1.8401412002017146,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004905703798752248,
|
|
"loss": 5.928,
|
|
"mean_token_accuracy": 0.1296511970460415,
|
|
"num_tokens": 4023690.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.838856601715088,
|
|
"epoch": 1.8451840645486637,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000490456413911464,
|
|
"loss": 5.7606,
|
|
"mean_token_accuracy": 0.13645405173301697,
|
|
"num_tokens": 4033338.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.848793077468872,
|
|
"epoch": 1.8502269288956126,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004903417783041987,
|
|
"loss": 5.7763,
|
|
"mean_token_accuracy": 0.1365301750600338,
|
|
"num_tokens": 4042873.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.928408908843994,
|
|
"epoch": 1.8552697932425617,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004902264734097188,
|
|
"loss": 5.9228,
|
|
"mean_token_accuracy": 0.13415781110525132,
|
|
"num_tokens": 4054049.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.961584663391113,
|
|
"epoch": 1.8603126575895108,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004901104995863948,
|
|
"loss": 5.9105,
|
|
"mean_token_accuracy": 0.12920038104057313,
|
|
"num_tokens": 4064254.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.959888362884522,
|
|
"epoch": 1.86535552193646,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004899938571946759,
|
|
"loss": 5.9088,
|
|
"mean_token_accuracy": 0.13008437752723695,
|
|
"num_tokens": 4075235.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.01555323600769,
|
|
"epoch": 1.870398386283409,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004898765465970894,
|
|
"loss": 6.0332,
|
|
"mean_token_accuracy": 0.12419131100177765,
|
|
"num_tokens": 4087434.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.92230715751648,
|
|
"epoch": 1.8754412506303582,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004897585681582393,
|
|
"loss": 5.8621,
|
|
"mean_token_accuracy": 0.1299331970512867,
|
|
"num_tokens": 4099223.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.86011061668396,
|
|
"epoch": 1.8804841149773073,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004896399222448053,
|
|
"loss": 5.8249,
|
|
"mean_token_accuracy": 0.1367386259138584,
|
|
"num_tokens": 4110706.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.009511661529541,
|
|
"epoch": 1.8855269793242562,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895206092255417,
|
|
"loss": 5.9138,
|
|
"mean_token_accuracy": 0.12570383846759797,
|
|
"num_tokens": 4122358.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.795359373092651,
|
|
"epoch": 1.8905698436712053,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004894006294712761,
|
|
"loss": 5.6944,
|
|
"mean_token_accuracy": 0.1380500726401806,
|
|
"num_tokens": 4132603.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.757789611816406,
|
|
"epoch": 1.8956127080181542,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004892799833549083,
|
|
"loss": 5.8502,
|
|
"mean_token_accuracy": 0.13927022740244865,
|
|
"num_tokens": 4144708.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.990426540374756,
|
|
"epoch": 1.9006555723651033,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004891586712514091,
|
|
"loss": 5.8796,
|
|
"mean_token_accuracy": 0.13205204233527185,
|
|
"num_tokens": 4155751.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.84541425704956,
|
|
"epoch": 1.9056984367120524,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004890366935378196,
|
|
"loss": 5.8613,
|
|
"mean_token_accuracy": 0.13200437948107718,
|
|
"num_tokens": 4167368.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.916485404968261,
|
|
"epoch": 1.9107413010590015,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004889140505932492,
|
|
"loss": 5.7987,
|
|
"mean_token_accuracy": 0.1378105029463768,
|
|
"num_tokens": 4178601.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.882160186767578,
|
|
"epoch": 1.9157841654059506,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004887907427988749,
|
|
"loss": 5.862,
|
|
"mean_token_accuracy": 0.12706216275691987,
|
|
"num_tokens": 4190421.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.8674720287322994,
|
|
"epoch": 1.9208270297528998,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004886667705379403,
|
|
"loss": 5.8175,
|
|
"mean_token_accuracy": 0.13355472385883332,
|
|
"num_tokens": 4201916.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.874330472946167,
|
|
"epoch": 1.9258698940998489,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004885421341957541,
|
|
"loss": 5.7994,
|
|
"mean_token_accuracy": 0.12998703494668007,
|
|
"num_tokens": 4213742.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.91501522064209,
|
|
"epoch": 1.9309127584467978,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004884168341596888,
|
|
"loss": 5.9826,
|
|
"mean_token_accuracy": 0.1286608800292015,
|
|
"num_tokens": 4226743.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.861303329467773,
|
|
"epoch": 1.9359556227937469,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004882908708191797,
|
|
"loss": 5.7969,
|
|
"mean_token_accuracy": 0.13505506440997123,
|
|
"num_tokens": 4237458.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.8801346778869625,
|
|
"epoch": 1.9409984871406958,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048816424456572403,
|
|
"loss": 5.853,
|
|
"mean_token_accuracy": 0.13410338014364243,
|
|
"num_tokens": 4248466.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.8861486434936525,
|
|
"epoch": 1.946041351487645,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048803695579287905,
|
|
"loss": 5.9155,
|
|
"mean_token_accuracy": 0.1292600877583027,
|
|
"num_tokens": 4260186.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.9890905857086185,
|
|
"epoch": 1.951084215834594,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048790900489626104,
|
|
"loss": 5.9427,
|
|
"mean_token_accuracy": 0.12938498109579086,
|
|
"num_tokens": 4272357.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.975081062316894,
|
|
"epoch": 1.9561270801815431,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004877803922735445,
|
|
"loss": 5.9479,
|
|
"mean_token_accuracy": 0.13041204810142518,
|
|
"num_tokens": 4285583.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.870516920089722,
|
|
"epoch": 1.9611699445284922,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004876511183244603,
|
|
"loss": 5.8324,
|
|
"mean_token_accuracy": 0.13399678468704224,
|
|
"num_tokens": 4296396.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.805631160736084,
|
|
"epoch": 1.9662128088754414,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048752118345079496,
|
|
"loss": 5.7974,
|
|
"mean_token_accuracy": 0.13942713886499405,
|
|
"num_tokens": 4306630.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.00920295715332,
|
|
"epoch": 1.9712556732223905,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048739058805638905,
|
|
"loss": 5.923,
|
|
"mean_token_accuracy": 0.1364424578845501,
|
|
"num_tokens": 4318488.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.812536239624023,
|
|
"epoch": 1.9762985375693394,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000487259332547136,
|
|
"loss": 5.7415,
|
|
"mean_token_accuracy": 0.13686809912323952,
|
|
"num_tokens": 4329604.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.856548261642456,
|
|
"epoch": 1.9813414019162885,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000487127417330981,
|
|
"loss": 5.7621,
|
|
"mean_token_accuracy": 0.13868005126714705,
|
|
"num_tokens": 4341325.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.849880933761597,
|
|
"epoch": 1.9863842662632374,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004869948428179196,
|
|
"loss": 5.8356,
|
|
"mean_token_accuracy": 0.1344212606549263,
|
|
"num_tokens": 4353308.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.881313610076904,
|
|
"epoch": 1.9914271306101865,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004868616094199964,
|
|
"loss": 5.7938,
|
|
"mean_token_accuracy": 0.13609838709235192,
|
|
"num_tokens": 4364937.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.780176305770874,
|
|
"epoch": 1.9964699949571356,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000486727717551304,
|
|
"loss": 5.7866,
|
|
"mean_token_accuracy": 0.13767975196242332,
|
|
"num_tokens": 4376477.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.966682434082031,
|
|
"epoch": 2.00100857286939,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004865931676279813,
|
|
"loss": 6.1222,
|
|
"mean_token_accuracy": 0.12126517958111233,
|
|
"num_tokens": 4386192.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.920079565048217,
|
|
"epoch": 2.006051437216339,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004864579600682125,
|
|
"loss": 5.5192,
|
|
"mean_token_accuracy": 0.14558632522821427,
|
|
"num_tokens": 4397170.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.766311836242676,
|
|
"epoch": 2.011094301563288,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048632209529222607,
|
|
"loss": 5.5758,
|
|
"mean_token_accuracy": 0.14178299009799958,
|
|
"num_tokens": 4407523.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.935072803497315,
|
|
"epoch": 2.016137165910237,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048618557372229253,
|
|
"loss": 5.5745,
|
|
"mean_token_accuracy": 0.13952894136309624,
|
|
"num_tokens": 4418947.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.016137165910237,
|
|
"eval_entropy": 5.7165008052311475,
|
|
"eval_loss": 5.974915027618408,
|
|
"eval_mean_token_accuracy": 0.13348384690648726,
|
|
"eval_num_tokens": 4418947.0,
|
|
"eval_runtime": 2.6409,
|
|
"eval_samples_per_second": 1188.245,
|
|
"eval_steps_per_second": 148.815,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.688140535354615,
|
|
"epoch": 2.0211800302571863,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004860483957827243,
|
|
"loss": 5.4905,
|
|
"mean_token_accuracy": 0.14712394177913665,
|
|
"num_tokens": 4429923.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.770442533493042,
|
|
"epoch": 2.026222894604135,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004859105618998735,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.14797610491514207,
|
|
"num_tokens": 4441490.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.791120147705078,
|
|
"epoch": 2.031265758951084,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004857720725021311,
|
|
"loss": 5.5229,
|
|
"mean_token_accuracy": 0.14146862626075746,
|
|
"num_tokens": 4453302.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.897823286056519,
|
|
"epoch": 2.036308623298033,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048563292801992534,
|
|
"loss": 5.4987,
|
|
"mean_token_accuracy": 0.1451796017587185,
|
|
"num_tokens": 4465393.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.624887275695801,
|
|
"epoch": 2.0413514876449823,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004854931288857203,
|
|
"loss": 5.4281,
|
|
"mean_token_accuracy": 0.15317499190568923,
|
|
"num_tokens": 4475269.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.740149974822998,
|
|
"epoch": 2.0463943519919314,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048535267553401505,
|
|
"loss": 5.4506,
|
|
"mean_token_accuracy": 0.15706779807806015,
|
|
"num_tokens": 4486275.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.748965501785278,
|
|
"epoch": 2.0514372163388805,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048521156840134193,
|
|
"loss": 5.4887,
|
|
"mean_token_accuracy": 0.14535945057868957,
|
|
"num_tokens": 4498684.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.789137077331543,
|
|
"epoch": 2.0564800806858297,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000485069807926265,
|
|
"loss": 5.5163,
|
|
"mean_token_accuracy": 0.14781803041696548,
|
|
"num_tokens": 4509482.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.7220855236053465,
|
|
"epoch": 2.0615229450327788,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004849273945493792,
|
|
"loss": 5.5873,
|
|
"mean_token_accuracy": 0.13608732670545579,
|
|
"num_tokens": 4520638.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.799525737762451,
|
|
"epoch": 2.066565809379728,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004847843287133084,
|
|
"loss": 5.596,
|
|
"mean_token_accuracy": 0.14232194870710374,
|
|
"num_tokens": 4531236.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.817762947082519,
|
|
"epoch": 2.0716086737266766,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004846406108627048,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.14913840889930724,
|
|
"num_tokens": 4542590.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.653972434997558,
|
|
"epoch": 2.0766515380736257,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048449624144424665,
|
|
"loss": 5.4833,
|
|
"mean_token_accuracy": 0.15156522691249846,
|
|
"num_tokens": 4553592.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.776318883895874,
|
|
"epoch": 2.081694402420575,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048435122090663744,
|
|
"loss": 5.5904,
|
|
"mean_token_accuracy": 0.1468003585934639,
|
|
"num_tokens": 4564735.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.825178718566894,
|
|
"epoch": 2.086737266767524,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048420554970060424,
|
|
"loss": 5.6439,
|
|
"mean_token_accuracy": 0.14915079921483992,
|
|
"num_tokens": 4576515.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.780494499206543,
|
|
"epoch": 2.091780131114473,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048405922827889677,
|
|
"loss": 5.5677,
|
|
"mean_token_accuracy": 0.1428247794508934,
|
|
"num_tokens": 4588212.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.887700366973877,
|
|
"epoch": 2.096822995461422,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048391225709628515,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.14353824853897096,
|
|
"num_tokens": 4599482.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.65536413192749,
|
|
"epoch": 2.1018658598083713,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004837646366095593,
|
|
"loss": 5.5566,
|
|
"mean_token_accuracy": 0.1518516346812248,
|
|
"num_tokens": 4609920.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.8641563892364506,
|
|
"epoch": 2.1069087241553204,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048361636727752716,
|
|
"loss": 5.6943,
|
|
"mean_token_accuracy": 0.13604221642017364,
|
|
"num_tokens": 4621944.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.724246883392334,
|
|
"epoch": 2.1119515885022695,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004834674495610131,
|
|
"loss": 5.564,
|
|
"mean_token_accuracy": 0.14400226026773452,
|
|
"num_tokens": 4634191.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.79607253074646,
|
|
"epoch": 2.116994452849218,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048331788392285693,
|
|
"loss": 5.5153,
|
|
"mean_token_accuracy": 0.14646561592817306,
|
|
"num_tokens": 4645655.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.688280439376831,
|
|
"epoch": 2.1220373171961673,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048316767082791205,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.14386310204863548,
|
|
"num_tokens": 4656922.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.762145566940307,
|
|
"epoch": 2.1270801815431164,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048301681074304427,
|
|
"loss": 5.5514,
|
|
"mean_token_accuracy": 0.14774558395147325,
|
|
"num_tokens": 4668647.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.7129593849182125,
|
|
"epoch": 2.1321230458900655,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048286530413713016,
|
|
"loss": 5.4646,
|
|
"mean_token_accuracy": 0.1492694079875946,
|
|
"num_tokens": 4678708.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.693505477905274,
|
|
"epoch": 2.1371659102370146,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048271315148105574,
|
|
"loss": 5.5788,
|
|
"mean_token_accuracy": 0.146945621073246,
|
|
"num_tokens": 4690962.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.746316862106323,
|
|
"epoch": 2.1422087745839637,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000482560353247715,
|
|
"loss": 5.6015,
|
|
"mean_token_accuracy": 0.14061713814735413,
|
|
"num_tokens": 4701833.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.674240255355835,
|
|
"epoch": 2.147251638930913,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004824069099120084,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.147596974670887,
|
|
"num_tokens": 4712756.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.7342997074127195,
|
|
"epoch": 2.152294503277862,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 5.5309,
|
|
"mean_token_accuracy": 0.14017585515975953,
|
|
"num_tokens": 4723226.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.725196743011475,
|
|
"epoch": 2.157337367624811,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004820980898431227,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.1423388034105301,
|
|
"num_tokens": 4734703.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.743356704711914,
|
|
"epoch": 2.1623802319717598,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004819427140697635,
|
|
"loss": 5.5509,
|
|
"mean_token_accuracy": 0.14523996859788896,
|
|
"num_tokens": 4745356.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.659573221206665,
|
|
"epoch": 2.167423096318709,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004817866951136753,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.1489861160516739,
|
|
"num_tokens": 4756342.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.754667043685913,
|
|
"epoch": 2.172465960665658,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048163003345976847,
|
|
"loss": 5.6071,
|
|
"mean_token_accuracy": 0.14500229954719543,
|
|
"num_tokens": 4768782.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.7073558330535885,
|
|
"epoch": 2.177508825012607,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004814727295949512,
|
|
"loss": 5.5012,
|
|
"mean_token_accuracy": 0.1480948284268379,
|
|
"num_tokens": 4779807.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.67652153968811,
|
|
"epoch": 2.1825516893595562,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048131478400812735,
|
|
"loss": 5.5534,
|
|
"mean_token_accuracy": 0.1416294828057289,
|
|
"num_tokens": 4789924.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.771294641494751,
|
|
"epoch": 2.1875945537065054,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004811561971901956,
|
|
"loss": 5.5901,
|
|
"mean_token_accuracy": 0.1433164209127426,
|
|
"num_tokens": 4801112.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.665855121612549,
|
|
"epoch": 2.1926374180534545,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004809969696340474,
|
|
"loss": 5.5258,
|
|
"mean_token_accuracy": 0.14302971139550208,
|
|
"num_tokens": 4811777.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.628685474395752,
|
|
"epoch": 2.1976802824004036,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048083710183456554,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.15123602449893953,
|
|
"num_tokens": 4822666.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.6726518154144285,
|
|
"epoch": 2.2027231467473527,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048067659428862306,
|
|
"loss": 5.4959,
|
|
"mean_token_accuracy": 0.1448014609515667,
|
|
"num_tokens": 4833315.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.700483512878418,
|
|
"epoch": 2.2077660110943014,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048051544749508096,
|
|
"loss": 5.552,
|
|
"mean_token_accuracy": 0.14341717287898065,
|
|
"num_tokens": 4844606.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.748077821731568,
|
|
"epoch": 2.2128088754412505,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004803536619547872,
|
|
"loss": 5.6252,
|
|
"mean_token_accuracy": 0.14131659865379334,
|
|
"num_tokens": 4857460.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.683910083770752,
|
|
"epoch": 2.2178517397881996,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000480191238170575,
|
|
"loss": 5.5277,
|
|
"mean_token_accuracy": 0.14796703606843947,
|
|
"num_tokens": 4869289.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.621657609939575,
|
|
"epoch": 2.2228946041351487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004800281766472612,
|
|
"loss": 5.5012,
|
|
"mean_token_accuracy": 0.14826309382915498,
|
|
"num_tokens": 4878912.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.725436496734619,
|
|
"epoch": 2.227937468482098,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047986447789164486,
|
|
"loss": 5.5777,
|
|
"mean_token_accuracy": 0.1396354019641876,
|
|
"num_tokens": 4891312.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.737303686141968,
|
|
"epoch": 2.232980332829047,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004797001424125054,
|
|
"loss": 5.585,
|
|
"mean_token_accuracy": 0.1401967354118824,
|
|
"num_tokens": 4902877.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.702682971954346,
|
|
"epoch": 2.238023197175996,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047953517072060124,
|
|
"loss": 5.4982,
|
|
"mean_token_accuracy": 0.1487305074930191,
|
|
"num_tokens": 4914145.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.738278722763061,
|
|
"epoch": 2.243066061522945,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004793695633286684,
|
|
"loss": 5.6342,
|
|
"mean_token_accuracy": 0.13958853334188462,
|
|
"num_tokens": 4924647.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.702555847167969,
|
|
"epoch": 2.2481089258698943,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047920332075141825,
|
|
"loss": 5.4958,
|
|
"mean_token_accuracy": 0.14522965103387833,
|
|
"num_tokens": 4934425.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.58786678314209,
|
|
"epoch": 2.253151790216843,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004790364435055367,
|
|
"loss": 5.4836,
|
|
"mean_token_accuracy": 0.15151820629835128,
|
|
"num_tokens": 4945045.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.675472831726074,
|
|
"epoch": 2.258194654563792,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000478868932109682,
|
|
"loss": 5.4835,
|
|
"mean_token_accuracy": 0.1467411383986473,
|
|
"num_tokens": 4956225.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.6820001125335695,
|
|
"epoch": 2.263237518910741,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004787007870844835,
|
|
"loss": 5.4893,
|
|
"mean_token_accuracy": 0.14946691393852235,
|
|
"num_tokens": 4968591.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.58260931968689,
|
|
"epoch": 2.2682803832576903,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047853200895253976,
|
|
"loss": 5.5583,
|
|
"mean_token_accuracy": 0.14588147550821304,
|
|
"num_tokens": 4978612.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.651156854629517,
|
|
"epoch": 2.2733232476046394,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047836259823841716,
|
|
"loss": 5.4441,
|
|
"mean_token_accuracy": 0.15017334818840028,
|
|
"num_tokens": 4990039.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.674918603897095,
|
|
"epoch": 2.2783661119515886,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047819255546864797,
|
|
"loss": 5.5703,
|
|
"mean_token_accuracy": 0.14660640731453894,
|
|
"num_tokens": 5001978.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.604819250106812,
|
|
"epoch": 2.2834089762985377,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004780218811717293,
|
|
"loss": 5.4722,
|
|
"mean_token_accuracy": 0.15239026844501496,
|
|
"num_tokens": 5012559.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.747910261154175,
|
|
"epoch": 2.288451840645487,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047785057587812057,
|
|
"loss": 5.5238,
|
|
"mean_token_accuracy": 0.14815680906176568,
|
|
"num_tokens": 5022944.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.67052116394043,
|
|
"epoch": 2.293494704992436,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047767864012024266,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.14561946988105773,
|
|
"num_tokens": 5033771.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.625165319442749,
|
|
"epoch": 2.2985375693393846,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047750607443247585,
|
|
"loss": 5.4905,
|
|
"mean_token_accuracy": 0.15097652673721312,
|
|
"num_tokens": 5045631.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.718620777130127,
|
|
"epoch": 2.3035804336863337,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047733287935115837,
|
|
"loss": 5.4859,
|
|
"mean_token_accuracy": 0.14836325496435165,
|
|
"num_tokens": 5056392.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.577760028839111,
|
|
"epoch": 2.308623298033283,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004771590554145844,
|
|
"loss": 5.5731,
|
|
"mean_token_accuracy": 0.15519963204860687,
|
|
"num_tokens": 5066420.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.729710531234741,
|
|
"epoch": 2.313666162380232,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047698460316300285,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.14488720446825026,
|
|
"num_tokens": 5075608.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.6740076541900635,
|
|
"epoch": 2.318709026727181,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004768095231386153,
|
|
"loss": 5.5265,
|
|
"mean_token_accuracy": 0.14924899190664292,
|
|
"num_tokens": 5086304.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.702730703353882,
|
|
"epoch": 2.32375189107413,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004766338158855744,
|
|
"loss": 5.5951,
|
|
"mean_token_accuracy": 0.14297810941934586,
|
|
"num_tokens": 5096811.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.710735511779785,
|
|
"epoch": 2.3287947554210793,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004764574819499826,
|
|
"loss": 5.5755,
|
|
"mean_token_accuracy": 0.14123538881540298,
|
|
"num_tokens": 5107118.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.738034582138061,
|
|
"epoch": 2.3338376197680284,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047628052187988955,
|
|
"loss": 5.5514,
|
|
"mean_token_accuracy": 0.15025267228484154,
|
|
"num_tokens": 5118362.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.643430137634278,
|
|
"epoch": 2.338880484114977,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004761029362252914,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.1514013096690178,
|
|
"num_tokens": 5128705.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.639211702346802,
|
|
"epoch": 2.343923348461926,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047592472553812856,
|
|
"loss": 5.4308,
|
|
"mean_token_accuracy": 0.15441965758800508,
|
|
"num_tokens": 5139952.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.707053899765015,
|
|
"epoch": 2.3489662128088753,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004757458903722837,
|
|
"loss": 5.6183,
|
|
"mean_token_accuracy": 0.13712134584784508,
|
|
"num_tokens": 5149453.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.68634819984436,
|
|
"epoch": 2.3540090771558244,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047556643128358075,
|
|
"loss": 5.6103,
|
|
"mean_token_accuracy": 0.14707150533795357,
|
|
"num_tokens": 5160951.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.704366636276245,
|
|
"epoch": 2.3590519415027735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047538634882978277,
|
|
"loss": 5.5608,
|
|
"mean_token_accuracy": 0.14414046555757523,
|
|
"num_tokens": 5171172.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.769184732437134,
|
|
"epoch": 2.3640948058497226,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047520564357059013,
|
|
"loss": 5.5922,
|
|
"mean_token_accuracy": 0.14316453635692597,
|
|
"num_tokens": 5183178.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.586590814590454,
|
|
"epoch": 2.3691376701966718,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004750243160676389,
|
|
"loss": 5.5244,
|
|
"mean_token_accuracy": 0.14793571829795837,
|
|
"num_tokens": 5193719.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.6622710704803465,
|
|
"epoch": 2.374180534543621,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004748423668844992,
|
|
"loss": 5.4408,
|
|
"mean_token_accuracy": 0.15502848774194716,
|
|
"num_tokens": 5203873.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.62037353515625,
|
|
"epoch": 2.37922339889057,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004746597965866733,
|
|
"loss": 5.5127,
|
|
"mean_token_accuracy": 0.1461717367172241,
|
|
"num_tokens": 5213682.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.600476837158203,
|
|
"epoch": 2.384266263237519,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004744766057415938,
|
|
"loss": 5.4398,
|
|
"mean_token_accuracy": 0.14920284003019332,
|
|
"num_tokens": 5224850.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.665679979324341,
|
|
"epoch": 2.389309127584468,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004742927949186222,
|
|
"loss": 5.5412,
|
|
"mean_token_accuracy": 0.15057540088891982,
|
|
"num_tokens": 5237256.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.6559959888458256,
|
|
"epoch": 2.394351991931417,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047410836468904664,
|
|
"loss": 5.5618,
|
|
"mean_token_accuracy": 0.14627649784088134,
|
|
"num_tokens": 5247638.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.632528686523438,
|
|
"epoch": 2.399394856278366,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047392331562608073,
|
|
"loss": 5.4998,
|
|
"mean_token_accuracy": 0.14702658206224442,
|
|
"num_tokens": 5259438.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.69215202331543,
|
|
"epoch": 2.404437720625315,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004737376483048609,
|
|
"loss": 5.4901,
|
|
"mean_token_accuracy": 0.1548854447901249,
|
|
"num_tokens": 5269809.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.587312126159668,
|
|
"epoch": 2.4094805849722642,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004735513633024458,
|
|
"loss": 5.5387,
|
|
"mean_token_accuracy": 0.1431642770767212,
|
|
"num_tokens": 5280710.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.765651845932007,
|
|
"epoch": 2.4145234493192134,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004733644611978134,
|
|
"loss": 5.5453,
|
|
"mean_token_accuracy": 0.14382847920060157,
|
|
"num_tokens": 5293162.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.609010028839111,
|
|
"epoch": 2.4195663136661625,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004731769425718597,
|
|
"loss": 5.5804,
|
|
"mean_token_accuracy": 0.14515646547079086,
|
|
"num_tokens": 5304510.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.6429084777832035,
|
|
"epoch": 2.4246091780131116,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000472988808007397,
|
|
"loss": 5.4373,
|
|
"mean_token_accuracy": 0.14908331930637359,
|
|
"num_tokens": 5314962.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.6114805221557615,
|
|
"epoch": 2.4296520423600603,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047280005808915176,
|
|
"loss": 5.5188,
|
|
"mean_token_accuracy": 0.14984937012195587,
|
|
"num_tokens": 5325314.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.680434942245483,
|
|
"epoch": 2.4346949067070094,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004726106934037631,
|
|
"loss": 5.5899,
|
|
"mean_token_accuracy": 0.14161864668130875,
|
|
"num_tokens": 5336408.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.650278377532959,
|
|
"epoch": 2.4397377710539585,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004724207145397809,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.15070291012525558,
|
|
"num_tokens": 5348014.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.612211799621582,
|
|
"epoch": 2.4447806354009076,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047223012208766384,
|
|
"loss": 5.4997,
|
|
"mean_token_accuracy": 0.15422628223896026,
|
|
"num_tokens": 5357275.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.655633640289307,
|
|
"epoch": 2.4498234997478567,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004720389166397777,
|
|
"loss": 5.4646,
|
|
"mean_token_accuracy": 0.14888779520988465,
|
|
"num_tokens": 5367822.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.659993982315063,
|
|
"epoch": 2.454866364094806,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047184709879039325,
|
|
"loss": 5.5216,
|
|
"mean_token_accuracy": 0.14660426527261733,
|
|
"num_tokens": 5379193.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.592276620864868,
|
|
"epoch": 2.459909228441755,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047165466913568497,
|
|
"loss": 5.4914,
|
|
"mean_token_accuracy": 0.1498927131295204,
|
|
"num_tokens": 5389825.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.517462110519409,
|
|
"epoch": 2.464952092788704,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004714616282737285,
|
|
"loss": 5.3926,
|
|
"mean_token_accuracy": 0.15678471475839614,
|
|
"num_tokens": 5400318.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.629178714752197,
|
|
"epoch": 2.469994957135653,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004712679768044995,
|
|
"loss": 5.5389,
|
|
"mean_token_accuracy": 0.1479016587138176,
|
|
"num_tokens": 5411432.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.653389549255371,
|
|
"epoch": 2.4750378214826023,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000471073715329871,
|
|
"loss": 5.5909,
|
|
"mean_token_accuracy": 0.14415307492017745,
|
|
"num_tokens": 5422662.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.606546401977539,
|
|
"epoch": 2.4800806858295514,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004708788444536123,
|
|
"loss": 5.6083,
|
|
"mean_token_accuracy": 0.14451165348291398,
|
|
"num_tokens": 5434976.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.709273767471314,
|
|
"epoch": 2.4851235501765,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004706833647813865,
|
|
"loss": 5.4878,
|
|
"mean_token_accuracy": 0.14486440569162368,
|
|
"num_tokens": 5445607.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.710936784744263,
|
|
"epoch": 2.4901664145234492,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047048727692074893,
|
|
"loss": 5.6434,
|
|
"mean_token_accuracy": 0.14222367703914643,
|
|
"num_tokens": 5455793.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.652678918838501,
|
|
"epoch": 2.4952092788703983,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004702905814811452,
|
|
"loss": 5.5881,
|
|
"mean_token_accuracy": 0.14239798486232758,
|
|
"num_tokens": 5466800.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.7010115623474125,
|
|
"epoch": 2.5002521432173475,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047009327907390926,
|
|
"loss": 5.5406,
|
|
"mean_token_accuracy": 0.14696398004889488,
|
|
"num_tokens": 5477875.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.590509605407715,
|
|
"epoch": 2.5052950075642966,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004698953703122616,
|
|
"loss": 5.5611,
|
|
"mean_token_accuracy": 0.14322107136249543,
|
|
"num_tokens": 5488193.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.782047271728516,
|
|
"epoch": 2.5103378719112457,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00046969685581130715,
|
|
"loss": 5.7244,
|
|
"mean_token_accuracy": 0.1373824082314968,
|
|
"num_tokens": 5500179.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.704003095626831,
|
|
"epoch": 2.515380736258195,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046949773618803353,
|
|
"loss": 5.5792,
|
|
"mean_token_accuracy": 0.13811678811907768,
|
|
"num_tokens": 5513413.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.694434595108032,
|
|
"epoch": 2.5204236006051435,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046929801206130925,
|
|
"loss": 5.5885,
|
|
"mean_token_accuracy": 0.14615954756736754,
|
|
"num_tokens": 5524697.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.5204236006051435,
|
|
"eval_entropy": 5.569981584718816,
|
|
"eval_loss": 5.916189670562744,
|
|
"eval_mean_token_accuracy": 0.13634676808543483,
|
|
"eval_num_tokens": 5524697.0,
|
|
"eval_runtime": 2.6358,
|
|
"eval_samples_per_second": 1190.523,
|
|
"eval_steps_per_second": 149.1,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.715534448623657,
|
|
"epoch": 2.5254664649520926,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046909768405188135,
|
|
"loss": 5.6376,
|
|
"mean_token_accuracy": 0.14564819484949112,
|
|
"num_tokens": 5536816.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.663525581359863,
|
|
"epoch": 2.5305093292990417,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00046889675278237404,
|
|
"loss": 5.6021,
|
|
"mean_token_accuracy": 0.14897046610713005,
|
|
"num_tokens": 5549329.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.640772485733033,
|
|
"epoch": 2.535552193645991,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004686952188772863,
|
|
"loss": 5.4746,
|
|
"mean_token_accuracy": 0.15271224081516266,
|
|
"num_tokens": 5560015.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.686579275131225,
|
|
"epoch": 2.54059505799294,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046849308296299005,
|
|
"loss": 5.5935,
|
|
"mean_token_accuracy": 0.14233456403017045,
|
|
"num_tokens": 5571031.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.681400966644287,
|
|
"epoch": 2.545637922339889,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004682903456677285,
|
|
"loss": 5.5679,
|
|
"mean_token_accuracy": 0.1482929654419422,
|
|
"num_tokens": 5582283.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.7204197406768795,
|
|
"epoch": 2.550680786686838,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046808700762161385,
|
|
"loss": 5.594,
|
|
"mean_token_accuracy": 0.14391592592000962,
|
|
"num_tokens": 5593185.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.6744321346282955,
|
|
"epoch": 2.5557236510337873,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004678830694566253,
|
|
"loss": 5.4978,
|
|
"mean_token_accuracy": 0.14866942316293716,
|
|
"num_tokens": 5604147.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.506406497955322,
|
|
"epoch": 2.5607665153807364,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 5.4899,
|
|
"mean_token_accuracy": 0.14699023813009263,
|
|
"num_tokens": 5615074.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.619188117980957,
|
|
"epoch": 2.5658093797276855,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004674733953072682,
|
|
"loss": 5.5366,
|
|
"mean_token_accuracy": 0.15007463097572327,
|
|
"num_tokens": 5626599.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.606451225280762,
|
|
"epoch": 2.5708522440746346,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004672676605961763,
|
|
"loss": 5.4654,
|
|
"mean_token_accuracy": 0.15293785482645034,
|
|
"num_tokens": 5637378.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.608837270736695,
|
|
"epoch": 2.5758951084215833,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046706132831276007,
|
|
"loss": 5.4957,
|
|
"mean_token_accuracy": 0.15240262746810912,
|
|
"num_tokens": 5649152.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.674561071395874,
|
|
"epoch": 2.5809379727685324,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046685439909830486,
|
|
"loss": 5.591,
|
|
"mean_token_accuracy": 0.1499859184026718,
|
|
"num_tokens": 5660476.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.571696519851685,
|
|
"epoch": 2.5859808371154815,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004666468735959516,
|
|
"loss": 5.4435,
|
|
"mean_token_accuracy": 0.152605539560318,
|
|
"num_tokens": 5672178.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.64791316986084,
|
|
"epoch": 2.5910237014624307,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004664387524506943,
|
|
"loss": 5.5385,
|
|
"mean_token_accuracy": 0.1452690616250038,
|
|
"num_tokens": 5683988.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.638162469863891,
|
|
"epoch": 2.59606656580938,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004662300363093784,
|
|
"loss": 5.566,
|
|
"mean_token_accuracy": 0.14803780764341354,
|
|
"num_tokens": 5695009.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.633869409561157,
|
|
"epoch": 2.601109430156329,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004660207258206983,
|
|
"loss": 5.556,
|
|
"mean_token_accuracy": 0.14445877969264984,
|
|
"num_tokens": 5706838.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.617272233963012,
|
|
"epoch": 2.606152294503278,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046581082163519585,
|
|
"loss": 5.448,
|
|
"mean_token_accuracy": 0.15211595892906188,
|
|
"num_tokens": 5717846.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.605248832702637,
|
|
"epoch": 2.6111951588502267,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004656003244052582,
|
|
"loss": 5.4754,
|
|
"mean_token_accuracy": 0.14720592200756072,
|
|
"num_tokens": 5727519.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.626052570343018,
|
|
"epoch": 2.616238023197176,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004653892347851155,
|
|
"loss": 5.5436,
|
|
"mean_token_accuracy": 0.1400951400399208,
|
|
"num_tokens": 5738717.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.694786405563354,
|
|
"epoch": 2.621280887544125,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004651775534308392,
|
|
"loss": 5.6016,
|
|
"mean_token_accuracy": 0.14594041481614112,
|
|
"num_tokens": 5750095.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.621736240386963,
|
|
"epoch": 2.626323751891074,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004649652810003398,
|
|
"loss": 5.4696,
|
|
"mean_token_accuracy": 0.15165551304817199,
|
|
"num_tokens": 5760282.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.6037664890289305,
|
|
"epoch": 2.631366616238023,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046475241815336504,
|
|
"loss": 5.5187,
|
|
"mean_token_accuracy": 0.14478351324796676,
|
|
"num_tokens": 5770547.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.631533622741699,
|
|
"epoch": 2.6364094805849723,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004645389655514974,
|
|
"loss": 5.4969,
|
|
"mean_token_accuracy": 0.14975763708353043,
|
|
"num_tokens": 5780801.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.684841108322144,
|
|
"epoch": 2.6414523449319214,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004643249238581525,
|
|
"loss": 5.5415,
|
|
"mean_token_accuracy": 0.14400465935468673,
|
|
"num_tokens": 5791851.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.636694002151489,
|
|
"epoch": 2.6464952092788705,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004641102937385769,
|
|
"loss": 5.6162,
|
|
"mean_token_accuracy": 0.1403758630156517,
|
|
"num_tokens": 5802551.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.620252275466919,
|
|
"epoch": 2.6515380736258196,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004638950758598461,
|
|
"loss": 5.4711,
|
|
"mean_token_accuracy": 0.1502656802535057,
|
|
"num_tokens": 5813376.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.625906467437744,
|
|
"epoch": 2.6565809379727687,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046367927089086203,
|
|
"loss": 5.5303,
|
|
"mean_token_accuracy": 0.14435117393732072,
|
|
"num_tokens": 5824634.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.620067501068116,
|
|
"epoch": 2.661623802319718,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00046346287950235163,
|
|
"loss": 5.4227,
|
|
"mean_token_accuracy": 0.15278890430927278,
|
|
"num_tokens": 5836771.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.5916036605834964,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004632459023668642,
|
|
"loss": 5.4494,
|
|
"mean_token_accuracy": 0.14611073359847068,
|
|
"num_tokens": 5847695.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.585438299179077,
|
|
"epoch": 2.6717095310136156,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004630283401587699,
|
|
"loss": 5.5338,
|
|
"mean_token_accuracy": 0.1470185749232769,
|
|
"num_tokens": 5859843.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.619991207122803,
|
|
"epoch": 2.6767523953605648,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004628101935542568,
|
|
"loss": 5.4991,
|
|
"mean_token_accuracy": 0.1494264855980873,
|
|
"num_tokens": 5871093.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.59155764579773,
|
|
"epoch": 2.681795259707514,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046259146323132986,
|
|
"loss": 5.4893,
|
|
"mean_token_accuracy": 0.15731683522462844,
|
|
"num_tokens": 5882235.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.544424819946289,
|
|
"epoch": 2.686838124054463,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046237214986980776,
|
|
"loss": 5.4143,
|
|
"mean_token_accuracy": 0.15443341434001923,
|
|
"num_tokens": 5892567.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.669776868820191,
|
|
"epoch": 2.691880988401412,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004621522541513216,
|
|
"loss": 5.5736,
|
|
"mean_token_accuracy": 0.14649525731801988,
|
|
"num_tokens": 5904849.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.6247350692749025,
|
|
"epoch": 2.696923852748361,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004619317767593123,
|
|
"loss": 5.5149,
|
|
"mean_token_accuracy": 0.14898416474461557,
|
|
"num_tokens": 5914873.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.647739696502685,
|
|
"epoch": 2.70196671709531,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046171071837902873,
|
|
"loss": 5.5413,
|
|
"mean_token_accuracy": 0.14528264254331588,
|
|
"num_tokens": 5926312.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.601176309585571,
|
|
"epoch": 2.707009581442259,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004614890796975254,
|
|
"loss": 5.5962,
|
|
"mean_token_accuracy": 0.14780106246471406,
|
|
"num_tokens": 5939665.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.626812934875488,
|
|
"epoch": 2.712052445789208,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004612668614036605,
|
|
"loss": 5.5046,
|
|
"mean_token_accuracy": 0.1570080816745758,
|
|
"num_tokens": 5949999.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.612789058685303,
|
|
"epoch": 2.7170953101361572,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004610440641880934,
|
|
"loss": 5.4787,
|
|
"mean_token_accuracy": 0.15323931947350503,
|
|
"num_tokens": 5960026.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.532758378982544,
|
|
"epoch": 2.7221381744831064,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046082068874328324,
|
|
"loss": 5.4822,
|
|
"mean_token_accuracy": 0.14454658031463624,
|
|
"num_tokens": 5971505.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.682231378555298,
|
|
"epoch": 2.7271810388300555,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046059673576348596,
|
|
"loss": 5.4588,
|
|
"mean_token_accuracy": 0.15356901586055755,
|
|
"num_tokens": 5982919.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.490727090835572,
|
|
"epoch": 2.7322239031770046,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046037220594475264,
|
|
"loss": 5.4787,
|
|
"mean_token_accuracy": 0.14964330568909645,
|
|
"num_tokens": 5993146.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.609921073913574,
|
|
"epoch": 2.7372667675239537,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000460147099984927,
|
|
"loss": 5.4831,
|
|
"mean_token_accuracy": 0.14875664785504342,
|
|
"num_tokens": 6004864.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.591419696807861,
|
|
"epoch": 2.742309631870903,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004599214185836438,
|
|
"loss": 5.4795,
|
|
"mean_token_accuracy": 0.14928718283772469,
|
|
"num_tokens": 6016440.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.523524332046509,
|
|
"epoch": 2.747352496217852,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045969516244232574,
|
|
"loss": 5.4372,
|
|
"mean_token_accuracy": 0.15902109295129777,
|
|
"num_tokens": 6026227.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.582226324081421,
|
|
"epoch": 2.752395360564801,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045946833226418235,
|
|
"loss": 5.499,
|
|
"mean_token_accuracy": 0.1450572945177555,
|
|
"num_tokens": 6038290.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.630072546005249,
|
|
"epoch": 2.7574382249117497,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004592409287542069,
|
|
"loss": 5.5257,
|
|
"mean_token_accuracy": 0.14881781116127968,
|
|
"num_tokens": 6049611.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.658566808700561,
|
|
"epoch": 2.762481089258699,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004590129526191748,
|
|
"loss": 5.5283,
|
|
"mean_token_accuracy": 0.15084161460399628,
|
|
"num_tokens": 6060270.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.582420682907104,
|
|
"epoch": 2.767523953605648,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004587844045676413,
|
|
"loss": 5.5306,
|
|
"mean_token_accuracy": 0.15279269218444824,
|
|
"num_tokens": 6071380.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.655093336105347,
|
|
"epoch": 2.772566817952597,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045855528530993874,
|
|
"loss": 5.564,
|
|
"mean_token_accuracy": 0.14139169007539748,
|
|
"num_tokens": 6082655.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.594664430618286,
|
|
"epoch": 2.777609682299546,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045832559555817525,
|
|
"loss": 5.4424,
|
|
"mean_token_accuracy": 0.1508428856730461,
|
|
"num_tokens": 6094446.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.65729866027832,
|
|
"epoch": 2.7826525466464953,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045809533602623185,
|
|
"loss": 5.5457,
|
|
"mean_token_accuracy": 0.14784732162952424,
|
|
"num_tokens": 6105606.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.543753385543823,
|
|
"epoch": 2.7876954109934444,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004578645074297603,
|
|
"loss": 5.5149,
|
|
"mean_token_accuracy": 0.1504577398300171,
|
|
"num_tokens": 6117743.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.611225271224976,
|
|
"epoch": 2.792738275340393,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045763311048618144,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.15163731351494789,
|
|
"num_tokens": 6128324.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.623370504379272,
|
|
"epoch": 2.797781139687342,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004574011459146822,
|
|
"loss": 5.5027,
|
|
"mean_token_accuracy": 0.14757978320121765,
|
|
"num_tokens": 6138742.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.694330739974975,
|
|
"epoch": 2.8028240040342913,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045716861443621386,
|
|
"loss": 5.5109,
|
|
"mean_token_accuracy": 0.14752026051282882,
|
|
"num_tokens": 6149990.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.572966194152832,
|
|
"epoch": 2.8078668683812404,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045693551677348967,
|
|
"loss": 5.5465,
|
|
"mean_token_accuracy": 0.1497933492064476,
|
|
"num_tokens": 6160281.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.601632356643677,
|
|
"epoch": 2.8129097327281896,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004567018536509823,
|
|
"loss": 5.5768,
|
|
"mean_token_accuracy": 0.1480606362223625,
|
|
"num_tokens": 6171501.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.7564770698547365,
|
|
"epoch": 2.8179525970751387,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045646762579492245,
|
|
"loss": 5.6462,
|
|
"mean_token_accuracy": 0.14339930117130278,
|
|
"num_tokens": 6183401.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.63318281173706,
|
|
"epoch": 2.822995461422088,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045623283393329546,
|
|
"loss": 5.4648,
|
|
"mean_token_accuracy": 0.15423868894577025,
|
|
"num_tokens": 6194451.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.533782958984375,
|
|
"epoch": 2.828038325769037,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045599747879584,
|
|
"loss": 5.4272,
|
|
"mean_token_accuracy": 0.1552419275045395,
|
|
"num_tokens": 6204410.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.581860542297363,
|
|
"epoch": 2.833081190115986,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004557615611140451,
|
|
"loss": 5.4663,
|
|
"mean_token_accuracy": 0.14584519416093827,
|
|
"num_tokens": 6215091.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.628701496124267,
|
|
"epoch": 2.838124054462935,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004555250816211485,
|
|
"loss": 5.4699,
|
|
"mean_token_accuracy": 0.1537807837128639,
|
|
"num_tokens": 6227201.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.635453462600708,
|
|
"epoch": 2.8431669188098843,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045528804105213395,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.1468923158943653,
|
|
"num_tokens": 6237067.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.607160139083862,
|
|
"epoch": 2.848209783156833,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004550504401437289,
|
|
"loss": 5.5175,
|
|
"mean_token_accuracy": 0.1566188246011734,
|
|
"num_tokens": 6247269.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.580532646179199,
|
|
"epoch": 2.853252647503782,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045481227963440235,
|
|
"loss": 5.5552,
|
|
"mean_token_accuracy": 0.14393058121204377,
|
|
"num_tokens": 6259094.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.662309265136718,
|
|
"epoch": 2.858295511850731,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045457356026436295,
|
|
"loss": 5.483,
|
|
"mean_token_accuracy": 0.1481974482536316,
|
|
"num_tokens": 6270012.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.528801679611206,
|
|
"epoch": 2.8633383761976803,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045433428277555577,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.15124562233686448,
|
|
"num_tokens": 6280936.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.617532968521118,
|
|
"epoch": 2.8683812405446294,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004540944479116609,
|
|
"loss": 5.4852,
|
|
"mean_token_accuracy": 0.15167278051376343,
|
|
"num_tokens": 6291471.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.591040420532226,
|
|
"epoch": 2.8734241048915785,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004538540564180906,
|
|
"loss": 5.5498,
|
|
"mean_token_accuracy": 0.1447784885764122,
|
|
"num_tokens": 6303717.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.659421777725219,
|
|
"epoch": 2.8784669692385276,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004536131090419871,
|
|
"loss": 5.5347,
|
|
"mean_token_accuracy": 0.14546699672937394,
|
|
"num_tokens": 6312960.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.690258121490478,
|
|
"epoch": 2.8835098335854763,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004533716065322205,
|
|
"loss": 5.5191,
|
|
"mean_token_accuracy": 0.14920947849750518,
|
|
"num_tokens": 6322558.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.709000444412231,
|
|
"epoch": 2.8885526979324254,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004531295496393861,
|
|
"loss": 5.6016,
|
|
"mean_token_accuracy": 0.14819440245628357,
|
|
"num_tokens": 6334763.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.636456823348999,
|
|
"epoch": 2.8935955622793745,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004528869391158023,
|
|
"loss": 5.5368,
|
|
"mean_token_accuracy": 0.14484109878540039,
|
|
"num_tokens": 6346678.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.617731189727783,
|
|
"epoch": 2.8986384266263236,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045264377571550834,
|
|
"loss": 5.5179,
|
|
"mean_token_accuracy": 0.15168764740228652,
|
|
"num_tokens": 6357495.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.540056657791138,
|
|
"epoch": 2.9036812909732728,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004524000601942615,
|
|
"loss": 5.4463,
|
|
"mean_token_accuracy": 0.15301640406250955,
|
|
"num_tokens": 6368557.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.5962952136993405,
|
|
"epoch": 2.908724155320222,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004521557933095354,
|
|
"loss": 5.4694,
|
|
"mean_token_accuracy": 0.15043903216719628,
|
|
"num_tokens": 6378471.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.568690586090088,
|
|
"epoch": 2.913767019667171,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000451910975820517,
|
|
"loss": 5.5524,
|
|
"mean_token_accuracy": 0.14270556718111038,
|
|
"num_tokens": 6390826.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.598607730865479,
|
|
"epoch": 2.91880988401412,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045166560848810476,
|
|
"loss": 5.3947,
|
|
"mean_token_accuracy": 0.15188634693622588,
|
|
"num_tokens": 6400438.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.525550794601441,
|
|
"epoch": 2.9238527483610692,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045141969207490606,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.15191902816295624,
|
|
"num_tokens": 6411533.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.56376576423645,
|
|
"epoch": 2.9288956127080183,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004511732273452347,
|
|
"loss": 5.4578,
|
|
"mean_token_accuracy": 0.15035423785448074,
|
|
"num_tokens": 6421409.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.606322145462036,
|
|
"epoch": 2.9339384770549675,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004509262150651089,
|
|
"loss": 5.5653,
|
|
"mean_token_accuracy": 0.1473022684454918,
|
|
"num_tokens": 6432738.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.606263065338135,
|
|
"epoch": 2.938981341401916,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045067865600224833,
|
|
"loss": 5.4981,
|
|
"mean_token_accuracy": 0.15135152339935304,
|
|
"num_tokens": 6443665.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.579765367507934,
|
|
"epoch": 2.9440242057488653,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004504305509260724,
|
|
"loss": 5.4426,
|
|
"mean_token_accuracy": 0.15822614431381227,
|
|
"num_tokens": 6454849.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.567010116577149,
|
|
"epoch": 2.9490670700958144,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004501819006076974,
|
|
"loss": 5.5083,
|
|
"mean_token_accuracy": 0.1529279351234436,
|
|
"num_tokens": 6465791.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.583162641525268,
|
|
"epoch": 2.9541099344427635,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044993270581993413,
|
|
"loss": 5.4388,
|
|
"mean_token_accuracy": 0.15790292471647263,
|
|
"num_tokens": 6475740.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.565184497833252,
|
|
"epoch": 2.9591527987897126,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00044968296733728587,
|
|
"loss": 5.48,
|
|
"mean_token_accuracy": 0.14736393392086028,
|
|
"num_tokens": 6487594.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.522630405426026,
|
|
"epoch": 2.9641956631366617,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004494326859359456,
|
|
"loss": 5.4349,
|
|
"mean_token_accuracy": 0.14542689472436904,
|
|
"num_tokens": 6498683.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.564045667648315,
|
|
"epoch": 2.969238527483611,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 5.4473,
|
|
"mean_token_accuracy": 0.14891347140073777,
|
|
"num_tokens": 6509069.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.619791841506958,
|
|
"epoch": 2.9742813918305595,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004489304974903953,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.14885881543159485,
|
|
"num_tokens": 6519480.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.550818204879761,
|
|
"epoch": 2.9793242561775086,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004486785920069985,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.15276284217834474,
|
|
"num_tokens": 6531010.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.6219322681427,
|
|
"epoch": 2.9843671205244577,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004484261467265313,
|
|
"loss": 5.4605,
|
|
"mean_token_accuracy": 0.14902912825345993,
|
|
"num_tokens": 6540292.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.664312839508057,
|
|
"epoch": 2.989409984871407,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044817316243359956,
|
|
"loss": 5.6143,
|
|
"mean_token_accuracy": 0.13999737054109573,
|
|
"num_tokens": 6553199.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.522039890289307,
|
|
"epoch": 2.994452849218356,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00044791963991448427,
|
|
"loss": 5.4789,
|
|
"mean_token_accuracy": 0.15009307116270065,
|
|
"num_tokens": 6564142.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.607178783416748,
|
|
"epoch": 2.999495713565305,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044766557995713915,
|
|
"loss": 5.4602,
|
|
"mean_token_accuracy": 0.15400594770908355,
|
|
"num_tokens": 6574519.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.563734372456868,
|
|
"epoch": 3.0040342914775593,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004474109833511885,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.1663739780584971,
|
|
"num_tokens": 6583348.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.530019807815552,
|
|
"epoch": 3.0090771558245084,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004471558508879244,
|
|
"loss": 5.2335,
|
|
"mean_token_accuracy": 0.16050369441509246,
|
|
"num_tokens": 6595704.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.53251576423645,
|
|
"epoch": 3.0141200201714575,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004469001833603045,
|
|
"loss": 5.1451,
|
|
"mean_token_accuracy": 0.16149887889623643,
|
|
"num_tokens": 6607177.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.47862663269043,
|
|
"epoch": 3.0191628845184066,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004466439815629493,
|
|
"loss": 5.1657,
|
|
"mean_token_accuracy": 0.1669058695435524,
|
|
"num_tokens": 6620798.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.5322123050689695,
|
|
"epoch": 3.0242057488653553,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044638724629214,
|
|
"loss": 5.2387,
|
|
"mean_token_accuracy": 0.16908339858055116,
|
|
"num_tokens": 6632972.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0242057488653553,
|
|
"eval_entropy": 5.518803171832446,
|
|
"eval_loss": 5.829952239990234,
|
|
"eval_mean_token_accuracy": 0.14320968427048383,
|
|
"eval_num_tokens": 6632972.0,
|
|
"eval_runtime": 2.8356,
|
|
"eval_samples_per_second": 1106.637,
|
|
"eval_steps_per_second": 138.594,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.498778438568115,
|
|
"epoch": 3.0292486132123044,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044612997834581557,
|
|
"loss": 5.0509,
|
|
"mean_token_accuracy": 0.1695736199617386,
|
|
"num_tokens": 6643259.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.455741310119629,
|
|
"epoch": 3.0342914775592535,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044587217852357086,
|
|
"loss": 5.2386,
|
|
"mean_token_accuracy": 0.16795216798782348,
|
|
"num_tokens": 6655875.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.57458815574646,
|
|
"epoch": 3.0393343419062027,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044561384762665363,
|
|
"loss": 5.1601,
|
|
"mean_token_accuracy": 0.16992995291948318,
|
|
"num_tokens": 6667377.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.408328199386597,
|
|
"epoch": 3.044377206253152,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004453549864579623,
|
|
"loss": 5.1415,
|
|
"mean_token_accuracy": 0.1619119167327881,
|
|
"num_tokens": 6678293.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.49166464805603,
|
|
"epoch": 3.049420070600101,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004450955958220433,
|
|
"loss": 5.1973,
|
|
"mean_token_accuracy": 0.16586864441633226,
|
|
"num_tokens": 6690664.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.666604042053223,
|
|
"epoch": 3.05446293494705,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044483567652508864,
|
|
"loss": 5.2032,
|
|
"mean_token_accuracy": 0.16186309605836868,
|
|
"num_tokens": 6701465.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.437854290008545,
|
|
"epoch": 3.059505799293999,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445752293749337,
|
|
"loss": 5.1995,
|
|
"mean_token_accuracy": 0.1668174758553505,
|
|
"num_tokens": 6713290.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.506089878082276,
|
|
"epoch": 3.0645486636409482,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004443142551810541,
|
|
"loss": 5.2168,
|
|
"mean_token_accuracy": 0.1657101944088936,
|
|
"num_tokens": 6723496.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.552100563049317,
|
|
"epoch": 3.069591527987897,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044405275475456365,
|
|
"loss": 5.1983,
|
|
"mean_token_accuracy": 0.163461734354496,
|
|
"num_tokens": 6734030.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.5378697395324705,
|
|
"epoch": 3.074634392334846,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004437907289082116,
|
|
"loss": 5.2287,
|
|
"mean_token_accuracy": 0.16727308332920074,
|
|
"num_tokens": 6745966.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.547027254104615,
|
|
"epoch": 3.079677256681795,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044352817845638053,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.16268666386604308,
|
|
"num_tokens": 6756982.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.420411109924316,
|
|
"epoch": 3.0847201210287443,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004432651042150833,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.16135234385728836,
|
|
"num_tokens": 6768950.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.402483034133911,
|
|
"epoch": 3.0897629853756934,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044300150700196054,
|
|
"loss": 5.1307,
|
|
"mean_token_accuracy": 0.15984653681516647,
|
|
"num_tokens": 6779748.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.5331212997436525,
|
|
"epoch": 3.0948058497226425,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004427373876362786,
|
|
"loss": 5.2637,
|
|
"mean_token_accuracy": 0.16265043020248413,
|
|
"num_tokens": 6791456.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.467827606201172,
|
|
"epoch": 3.0998487140695916,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004424727469389265,
|
|
"loss": 5.1834,
|
|
"mean_token_accuracy": 0.16567315012216569,
|
|
"num_tokens": 6802813.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.574123048782349,
|
|
"epoch": 3.1048915784165407,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004422075857324137,
|
|
"loss": 5.2427,
|
|
"mean_token_accuracy": 0.16068820655345917,
|
|
"num_tokens": 6813310.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.521329307556153,
|
|
"epoch": 3.10993444276349,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004419419048408673,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.16046479642391204,
|
|
"num_tokens": 6825593.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.441070795059204,
|
|
"epoch": 3.1149773071104385,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004416757050900297,
|
|
"loss": 5.1642,
|
|
"mean_token_accuracy": 0.16903398483991622,
|
|
"num_tokens": 6836279.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.441403913497925,
|
|
"epoch": 3.1200201714573876,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004414089873072557,
|
|
"loss": 5.1424,
|
|
"mean_token_accuracy": 0.16970179229974747,
|
|
"num_tokens": 6846087.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.491785383224487,
|
|
"epoch": 3.1250630358043368,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004411417523215105,
|
|
"loss": 5.1299,
|
|
"mean_token_accuracy": 0.1669689252972603,
|
|
"num_tokens": 6857290.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.386426877975464,
|
|
"epoch": 3.130105900151286,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004408740009633665,
|
|
"loss": 5.2361,
|
|
"mean_token_accuracy": 0.16384147852659225,
|
|
"num_tokens": 6870060.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.56688084602356,
|
|
"epoch": 3.135148764498235,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044060573406500117,
|
|
"loss": 5.1397,
|
|
"mean_token_accuracy": 0.15771555602550508,
|
|
"num_tokens": 6880813.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.425288105010987,
|
|
"epoch": 3.140191628845184,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004403369524601941,
|
|
"loss": 5.1509,
|
|
"mean_token_accuracy": 0.17193858474493026,
|
|
"num_tokens": 6891955.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.349575519561768,
|
|
"epoch": 3.145234493192133,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004400676569843249,
|
|
"loss": 5.1303,
|
|
"mean_token_accuracy": 0.16631903350353242,
|
|
"num_tokens": 6902873.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.5210484027862545,
|
|
"epoch": 3.1502773575390823,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00043979784847436994,
|
|
"loss": 5.1483,
|
|
"mean_token_accuracy": 0.16382198929786682,
|
|
"num_tokens": 6916104.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.48253321647644,
|
|
"epoch": 3.1553202218860315,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004395275277689004,
|
|
"loss": 5.1654,
|
|
"mean_token_accuracy": 0.16915008276700974,
|
|
"num_tokens": 6926117.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.411616468429566,
|
|
"epoch": 3.16036308623298,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004392566957080793,
|
|
"loss": 5.1602,
|
|
"mean_token_accuracy": 0.16095373928546905,
|
|
"num_tokens": 6936257.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.496952152252197,
|
|
"epoch": 3.1654059505799292,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043898535313365884,
|
|
"loss": 5.2959,
|
|
"mean_token_accuracy": 0.15923235267400743,
|
|
"num_tokens": 6948161.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.4647314071655275,
|
|
"epoch": 3.1704488149268784,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000438713500888978,
|
|
"loss": 5.1755,
|
|
"mean_token_accuracy": 0.16748863309621811,
|
|
"num_tokens": 6959014.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.465006303787232,
|
|
"epoch": 3.1754916792738275,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004384411398189599,
|
|
"loss": 5.1443,
|
|
"mean_token_accuracy": 0.16520180255174638,
|
|
"num_tokens": 6969180.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.440118408203125,
|
|
"epoch": 3.1805345436207766,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000438168270770109,
|
|
"loss": 5.2817,
|
|
"mean_token_accuracy": 0.16402920633554458,
|
|
"num_tokens": 6980702.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.530990362167358,
|
|
"epoch": 3.1855774079677257,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004378948945905085,
|
|
"loss": 5.2769,
|
|
"mean_token_accuracy": 0.15514416843652726,
|
|
"num_tokens": 6992904.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.461986017227173,
|
|
"epoch": 3.190620272314675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043762101212981805,
|
|
"loss": 5.166,
|
|
"mean_token_accuracy": 0.16778829246759414,
|
|
"num_tokens": 7004355.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.458444547653198,
|
|
"epoch": 3.195663136661624,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043734662423927054,
|
|
"loss": 5.1667,
|
|
"mean_token_accuracy": 0.1590945839881897,
|
|
"num_tokens": 7014941.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.393497896194458,
|
|
"epoch": 3.200706001008573,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004370717317716699,
|
|
"loss": 5.2185,
|
|
"mean_token_accuracy": 0.16269678771495819,
|
|
"num_tokens": 7026208.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.51782603263855,
|
|
"epoch": 3.205748865355522,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004367963355813882,
|
|
"loss": 5.2864,
|
|
"mean_token_accuracy": 0.1580560639500618,
|
|
"num_tokens": 7038065.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.545170831680298,
|
|
"epoch": 3.210791729702471,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043652043652436323,
|
|
"loss": 5.2436,
|
|
"mean_token_accuracy": 0.16035007685422897,
|
|
"num_tokens": 7049887.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.369115400314331,
|
|
"epoch": 3.21583459404942,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004362440354580956,
|
|
"loss": 5.1617,
|
|
"mean_token_accuracy": 0.1672530859708786,
|
|
"num_tokens": 7060056.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.413434743881226,
|
|
"epoch": 3.220877458396369,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004359671332416462,
|
|
"loss": 5.1881,
|
|
"mean_token_accuracy": 0.16316429674625396,
|
|
"num_tokens": 7071979.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.442484617233276,
|
|
"epoch": 3.225920322743318,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004356897307356335,
|
|
"loss": 5.2204,
|
|
"mean_token_accuracy": 0.16305021792650223,
|
|
"num_tokens": 7082534.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.467128324508667,
|
|
"epoch": 3.2309631870902673,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004354118288022309,
|
|
"loss": 5.2185,
|
|
"mean_token_accuracy": 0.16607303470373153,
|
|
"num_tokens": 7094737.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.457351493835449,
|
|
"epoch": 3.2360060514372164,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043513342830516405,
|
|
"loss": 5.2306,
|
|
"mean_token_accuracy": 0.16553484052419662,
|
|
"num_tokens": 7105483.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.41800708770752,
|
|
"epoch": 3.2410489157841655,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000434854530109708,
|
|
"loss": 5.1204,
|
|
"mean_token_accuracy": 0.16772897243499757,
|
|
"num_tokens": 7115110.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.467194128036499,
|
|
"epoch": 3.2460917801311147,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043457513508268487,
|
|
"loss": 5.2007,
|
|
"mean_token_accuracy": 0.16288117617368697,
|
|
"num_tokens": 7125216.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.4844239234924315,
|
|
"epoch": 3.2511346444780633,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043429524409246083,
|
|
"loss": 5.2418,
|
|
"mean_token_accuracy": 0.16317311972379683,
|
|
"num_tokens": 7136457.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.3839198589324955,
|
|
"epoch": 3.2561775088250124,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004340148580089436,
|
|
"loss": 5.218,
|
|
"mean_token_accuracy": 0.16534755378961563,
|
|
"num_tokens": 7147315.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.501992416381836,
|
|
"epoch": 3.2612203731719616,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004337339777035794,
|
|
"loss": 5.1912,
|
|
"mean_token_accuracy": 0.1601543202996254,
|
|
"num_tokens": 7158436.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.515243721008301,
|
|
"epoch": 3.2662632375189107,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000433452604049351,
|
|
"loss": 5.2561,
|
|
"mean_token_accuracy": 0.15643265545368196,
|
|
"num_tokens": 7168705.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.427344894409179,
|
|
"epoch": 3.27130610186586,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043317073792077394,
|
|
"loss": 5.1589,
|
|
"mean_token_accuracy": 0.16633531004190444,
|
|
"num_tokens": 7179200.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.479598760604858,
|
|
"epoch": 3.276348966212809,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004328883801938947,
|
|
"loss": 5.2449,
|
|
"mean_token_accuracy": 0.15928739458322524,
|
|
"num_tokens": 7189378.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.471361541748047,
|
|
"epoch": 3.281391830559758,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043260553174628776,
|
|
"loss": 5.1298,
|
|
"mean_token_accuracy": 0.17073576748371125,
|
|
"num_tokens": 7198726.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.450024509429932,
|
|
"epoch": 3.286434694906707,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004323221934570524,
|
|
"loss": 5.2056,
|
|
"mean_token_accuracy": 0.16591388285160064,
|
|
"num_tokens": 7210547.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.345797157287597,
|
|
"epoch": 3.2914775592536563,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043203836620681074,
|
|
"loss": 5.1179,
|
|
"mean_token_accuracy": 0.17085089385509492,
|
|
"num_tokens": 7221871.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.473985004425049,
|
|
"epoch": 3.2965204236006054,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043175405087770435,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.16670676320791245,
|
|
"num_tokens": 7233928.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.407733964920044,
|
|
"epoch": 3.301563287947554,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043146924835339186,
|
|
"loss": 5.2319,
|
|
"mean_token_accuracy": 0.1593266710639,
|
|
"num_tokens": 7245263.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.585802793502808,
|
|
"epoch": 3.306606152294503,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00043118395951904605,
|
|
"loss": 5.291,
|
|
"mean_token_accuracy": 0.16548178642988204,
|
|
"num_tokens": 7257941.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.458014726638794,
|
|
"epoch": 3.3116490166414523,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004308981852613512,
|
|
"loss": 5.1987,
|
|
"mean_token_accuracy": 0.16094456762075424,
|
|
"num_tokens": 7267854.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.418560647964478,
|
|
"epoch": 3.3166918809884014,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004306119264685003,
|
|
"loss": 5.2602,
|
|
"mean_token_accuracy": 0.16347412019968033,
|
|
"num_tokens": 7279898.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.524970197677613,
|
|
"epoch": 3.3217347453353505,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004303251840301923,
|
|
"loss": 5.2706,
|
|
"mean_token_accuracy": 0.1569972738623619,
|
|
"num_tokens": 7291459.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.534268474578857,
|
|
"epoch": 3.3267776096822996,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043003795883762933,
|
|
"loss": 5.2914,
|
|
"mean_token_accuracy": 0.16196816489100457,
|
|
"num_tokens": 7302709.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.408970832824707,
|
|
"epoch": 3.3318204740292487,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042975025178351386,
|
|
"loss": 5.1981,
|
|
"mean_token_accuracy": 0.1602844253182411,
|
|
"num_tokens": 7312541.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.445866823196411,
|
|
"epoch": 3.336863338376198,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000429462063762046,
|
|
"loss": 5.2243,
|
|
"mean_token_accuracy": 0.1612634390592575,
|
|
"num_tokens": 7323444.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.453121614456177,
|
|
"epoch": 3.3419062027231465,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042917339566892095,
|
|
"loss": 5.1671,
|
|
"mean_token_accuracy": 0.1690024495124817,
|
|
"num_tokens": 7334751.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.4250325679779055,
|
|
"epoch": 3.3469490670700957,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004288842484013257,
|
|
"loss": 5.259,
|
|
"mean_token_accuracy": 0.15966995507478715,
|
|
"num_tokens": 7347434.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.454699468612671,
|
|
"epoch": 3.3519919314170448,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004285946228579365,
|
|
"loss": 5.2701,
|
|
"mean_token_accuracy": 0.1651109591126442,
|
|
"num_tokens": 7358240.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.488889122009278,
|
|
"epoch": 3.357034795763994,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004283045199389164,
|
|
"loss": 5.237,
|
|
"mean_token_accuracy": 0.1610963061451912,
|
|
"num_tokens": 7369542.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.450783538818359,
|
|
"epoch": 3.362077660110943,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00042801394054591197,
|
|
"loss": 5.2791,
|
|
"mean_token_accuracy": 0.1544570133090019,
|
|
"num_tokens": 7381885.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.425880765914917,
|
|
"epoch": 3.367120524457892,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042772288558205057,
|
|
"loss": 5.1737,
|
|
"mean_token_accuracy": 0.16952318847179412,
|
|
"num_tokens": 7392042.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.509827041625977,
|
|
"epoch": 3.3721633888048412,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004274313559519377,
|
|
"loss": 5.2513,
|
|
"mean_token_accuracy": 0.16106278598308563,
|
|
"num_tokens": 7403388.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.423347806930542,
|
|
"epoch": 3.3772062531517903,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 5.1632,
|
|
"mean_token_accuracy": 0.16514763683080674,
|
|
"num_tokens": 7413501.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.396876478195191,
|
|
"epoch": 3.3822491174987395,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00042684687631875377,
|
|
"loss": 5.0678,
|
|
"mean_token_accuracy": 0.17176373898983002,
|
|
"num_tokens": 7424529.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.384559059143067,
|
|
"epoch": 3.3872919818456886,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00042655392813225864,
|
|
"loss": 5.242,
|
|
"mean_token_accuracy": 0.16137355268001558,
|
|
"num_tokens": 7434551.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.456484985351563,
|
|
"epoch": 3.3923348461926373,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042626050891265884,
|
|
"loss": 5.2433,
|
|
"mean_token_accuracy": 0.16101209819316864,
|
|
"num_tokens": 7443898.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.458910989761352,
|
|
"epoch": 3.3973777105395864,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004259666195719081,
|
|
"loss": 5.2172,
|
|
"mean_token_accuracy": 0.1675860896706581,
|
|
"num_tokens": 7454399.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.471589994430542,
|
|
"epoch": 3.4024205748865355,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042567226102342134,
|
|
"loss": 5.2639,
|
|
"mean_token_accuracy": 0.16877782642841338,
|
|
"num_tokens": 7465564.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.4233544826507565,
|
|
"epoch": 3.4074634392334846,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004253774341820715,
|
|
"loss": 5.2163,
|
|
"mean_token_accuracy": 0.1603696510195732,
|
|
"num_tokens": 7476619.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.406573724746704,
|
|
"epoch": 3.4125063035804337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004250821399641874,
|
|
"loss": 5.1855,
|
|
"mean_token_accuracy": 0.16541039049625397,
|
|
"num_tokens": 7487175.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.4396261215209964,
|
|
"epoch": 3.417549167927383,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004247863792875502,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.16160834431648255,
|
|
"num_tokens": 7498152.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.5115687370300295,
|
|
"epoch": 3.422592032274332,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042449015307139076,
|
|
"loss": 5.1836,
|
|
"mean_token_accuracy": 0.17419951260089875,
|
|
"num_tokens": 7507914.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.383336353302002,
|
|
"epoch": 3.427634896621281,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042419346223638725,
|
|
"loss": 5.2856,
|
|
"mean_token_accuracy": 0.1578053817152977,
|
|
"num_tokens": 7518941.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.3913439273834225,
|
|
"epoch": 3.4326777609682297,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004238963077046616,
|
|
"loss": 5.1124,
|
|
"mean_token_accuracy": 0.16630617678165435,
|
|
"num_tokens": 7529315.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.428023338317871,
|
|
"epoch": 3.437720625315179,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004235986903997767,
|
|
"loss": 5.2075,
|
|
"mean_token_accuracy": 0.1669050171971321,
|
|
"num_tokens": 7539888.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.478060293197632,
|
|
"epoch": 3.442763489662128,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042330061124673437,
|
|
"loss": 5.1955,
|
|
"mean_token_accuracy": 0.16277887374162675,
|
|
"num_tokens": 7550429.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.441017913818359,
|
|
"epoch": 3.447806354009077,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042300207117197116,
|
|
"loss": 5.1788,
|
|
"mean_token_accuracy": 0.16205011457204818,
|
|
"num_tokens": 7561051.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.431795597076416,
|
|
"epoch": 3.452849218356026,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042270307110335657,
|
|
"loss": 5.189,
|
|
"mean_token_accuracy": 0.16427184343338014,
|
|
"num_tokens": 7570847.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.370966148376465,
|
|
"epoch": 3.4578920827029753,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004224036119701899,
|
|
"loss": 5.1816,
|
|
"mean_token_accuracy": 0.16554466038942336,
|
|
"num_tokens": 7581515.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.356352710723877,
|
|
"epoch": 3.4629349470499244,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00042210369470319694,
|
|
"loss": 5.1995,
|
|
"mean_token_accuracy": 0.16370873153209686,
|
|
"num_tokens": 7592306.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.504908418655395,
|
|
"epoch": 3.4679778113968736,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004218033202345274,
|
|
"loss": 5.301,
|
|
"mean_token_accuracy": 0.15796938687562942,
|
|
"num_tokens": 7603264.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.465123796463013,
|
|
"epoch": 3.4730206757438227,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004215024894977522,
|
|
"loss": 5.2634,
|
|
"mean_token_accuracy": 0.16270803660154343,
|
|
"num_tokens": 7615603.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.411271572113037,
|
|
"epoch": 3.478063540090772,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004212012034278601,
|
|
"loss": 5.1508,
|
|
"mean_token_accuracy": 0.17226357161998748,
|
|
"num_tokens": 7626072.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.46341233253479,
|
|
"epoch": 3.4831064044377205,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042089946296125513,
|
|
"loss": 5.2184,
|
|
"mean_token_accuracy": 0.16225174367427825,
|
|
"num_tokens": 7636321.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.427269840240479,
|
|
"epoch": 3.4881492687846696,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042059726903575373,
|
|
"loss": 5.1926,
|
|
"mean_token_accuracy": 0.16928152441978456,
|
|
"num_tokens": 7647709.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.445364999771118,
|
|
"epoch": 3.4931921331316187,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004202946225905815,
|
|
"loss": 5.1903,
|
|
"mean_token_accuracy": 0.16657970994710922,
|
|
"num_tokens": 7660195.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.401914834976196,
|
|
"epoch": 3.498234997478568,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004199915245663706,
|
|
"loss": 5.146,
|
|
"mean_token_accuracy": 0.17144086360931396,
|
|
"num_tokens": 7670274.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.493826055526734,
|
|
"epoch": 3.503277861825517,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041968797590515673,
|
|
"loss": 5.2115,
|
|
"mean_token_accuracy": 0.161667200922966,
|
|
"num_tokens": 7680327.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.383252143859863,
|
|
"epoch": 3.508320726172466,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004193839775503761,
|
|
"loss": 5.2211,
|
|
"mean_token_accuracy": 0.16607730090618134,
|
|
"num_tokens": 7691250.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.408378458023071,
|
|
"epoch": 3.513363590519415,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00041907953044686253,
|
|
"loss": 5.2517,
|
|
"mean_token_accuracy": 0.16372339129447938,
|
|
"num_tokens": 7703726.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.4567506313323975,
|
|
"epoch": 3.518406454866364,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004187746355408448,
|
|
"loss": 5.2795,
|
|
"mean_token_accuracy": 0.16365584284067153,
|
|
"num_tokens": 7715179.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.508245992660522,
|
|
"epoch": 3.523449319213313,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004184692937799431,
|
|
"loss": 5.2743,
|
|
"mean_token_accuracy": 0.16309838443994523,
|
|
"num_tokens": 7726586.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.409369802474975,
|
|
"epoch": 3.528492183560262,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041816350611316693,
|
|
"loss": 5.2116,
|
|
"mean_token_accuracy": 0.1633657142519951,
|
|
"num_tokens": 7737350.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.528492183560262,
|
|
"eval_entropy": 5.288684311107218,
|
|
"eval_loss": 5.805794715881348,
|
|
"eval_mean_token_accuracy": 0.14353060472102566,
|
|
"eval_num_tokens": 7737350.0,
|
|
"eval_runtime": 2.639,
|
|
"eval_samples_per_second": 1189.077,
|
|
"eval_steps_per_second": 148.919,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.396514368057251,
|
|
"epoch": 3.533535047907211,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041785727349091117,
|
|
"loss": 5.2034,
|
|
"mean_token_accuracy": 0.16831117123365402,
|
|
"num_tokens": 7749586.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.383462715148926,
|
|
"epoch": 3.5385779122541603,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000417550596864954,
|
|
"loss": 5.2043,
|
|
"mean_token_accuracy": 0.1685790702700615,
|
|
"num_tokens": 7761889.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.432257461547851,
|
|
"epoch": 3.5436207766011094,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041724347718845344,
|
|
"loss": 5.2018,
|
|
"mean_token_accuracy": 0.16528122276067733,
|
|
"num_tokens": 7772088.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.431580591201782,
|
|
"epoch": 3.5486636409480585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041693591541594455,
|
|
"loss": 5.2121,
|
|
"mean_token_accuracy": 0.16515803933143616,
|
|
"num_tokens": 7783087.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.400111246109009,
|
|
"epoch": 3.5537065052950076,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004166279125033363,
|
|
"loss": 5.1533,
|
|
"mean_token_accuracy": 0.16039139032363892,
|
|
"num_tokens": 7792919.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.350706243515015,
|
|
"epoch": 3.5587493696419568,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00041631946940790893,
|
|
"loss": 5.1893,
|
|
"mean_token_accuracy": 0.16541917324066163,
|
|
"num_tokens": 7803905.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.476314353942871,
|
|
"epoch": 3.563792233988906,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041601058708831077,
|
|
"loss": 5.2159,
|
|
"mean_token_accuracy": 0.16536158919334412,
|
|
"num_tokens": 7815651.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.398756647109986,
|
|
"epoch": 3.568835098335855,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000415701266504555,
|
|
"loss": 5.1996,
|
|
"mean_token_accuracy": 0.16630251854658126,
|
|
"num_tokens": 7827247.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.432639455795288,
|
|
"epoch": 3.5738779626828037,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004153915086180173,
|
|
"loss": 5.2273,
|
|
"mean_token_accuracy": 0.16272253841161727,
|
|
"num_tokens": 7838461.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.353623676300049,
|
|
"epoch": 3.578920827029753,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004150813143914323,
|
|
"loss": 5.1592,
|
|
"mean_token_accuracy": 0.16748233139514923,
|
|
"num_tokens": 7849824.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.3674060821533205,
|
|
"epoch": 3.583963691376702,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004147706847888908,
|
|
"loss": 5.249,
|
|
"mean_token_accuracy": 0.16401712149381636,
|
|
"num_tokens": 7862167.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.460422658920288,
|
|
"epoch": 3.589006555723651,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004144596207758367,
|
|
"loss": 5.2182,
|
|
"mean_token_accuracy": 0.164128115773201,
|
|
"num_tokens": 7872320.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.418314218521118,
|
|
"epoch": 3.5940494200706,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004141481233190642,
|
|
"loss": 5.1856,
|
|
"mean_token_accuracy": 0.17166522443294524,
|
|
"num_tokens": 7883150.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.3734376430511475,
|
|
"epoch": 3.5990922844175492,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004138361933867145,
|
|
"loss": 5.2621,
|
|
"mean_token_accuracy": 0.15922135561704637,
|
|
"num_tokens": 7895371.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.366262292861938,
|
|
"epoch": 3.6041351487644984,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004135238319482731,
|
|
"loss": 5.1199,
|
|
"mean_token_accuracy": 0.17200247198343277,
|
|
"num_tokens": 7905669.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.384713935852051,
|
|
"epoch": 3.609178013111447,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004132110399745666,
|
|
"loss": 5.14,
|
|
"mean_token_accuracy": 0.16676151305437087,
|
|
"num_tokens": 7915714.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.372761964797974,
|
|
"epoch": 3.614220877458396,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004128978184377596,
|
|
"loss": 5.2178,
|
|
"mean_token_accuracy": 0.16233056485652925,
|
|
"num_tokens": 7927110.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.463270425796509,
|
|
"epoch": 3.6192637418053453,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004125841683113518,
|
|
"loss": 5.2482,
|
|
"mean_token_accuracy": 0.1620067447423935,
|
|
"num_tokens": 7939170.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.435369348526001,
|
|
"epoch": 3.6243066061522944,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00041227009057017523,
|
|
"loss": 5.2776,
|
|
"mean_token_accuracy": 0.15749395340681077,
|
|
"num_tokens": 7952615.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.4100950241088865,
|
|
"epoch": 3.6293494704992435,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004119555861903906,
|
|
"loss": 5.2362,
|
|
"mean_token_accuracy": 0.1592637062072754,
|
|
"num_tokens": 7963946.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.429463863372803,
|
|
"epoch": 3.6343923348461926,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004116406561494847,
|
|
"loss": 5.1996,
|
|
"mean_token_accuracy": 0.1630503699183464,
|
|
"num_tokens": 7975270.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.3208833694458,
|
|
"epoch": 3.6394351991931417,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041132530142626756,
|
|
"loss": 5.1475,
|
|
"mean_token_accuracy": 0.16887623071670532,
|
|
"num_tokens": 7986008.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.435577297210694,
|
|
"epoch": 3.644478063540091,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00041100952300086884,
|
|
"loss": 5.3155,
|
|
"mean_token_accuracy": 0.16070480048656463,
|
|
"num_tokens": 7997416.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.420179557800293,
|
|
"epoch": 3.64952092788704,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00041069332185473514,
|
|
"loss": 5.1047,
|
|
"mean_token_accuracy": 0.17083795964717866,
|
|
"num_tokens": 8009120.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.393123483657837,
|
|
"epoch": 3.654563792233989,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000410376698970627,
|
|
"loss": 5.225,
|
|
"mean_token_accuracy": 0.16107127219438552,
|
|
"num_tokens": 8020325.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.374727487564087,
|
|
"epoch": 3.659606656580938,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041005965533261554,
|
|
"loss": 5.1692,
|
|
"mean_token_accuracy": 0.16224971562623977,
|
|
"num_tokens": 8031456.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.425358057022095,
|
|
"epoch": 3.6646495209278873,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040974219192607973,
|
|
"loss": 5.2134,
|
|
"mean_token_accuracy": 0.16621486246585845,
|
|
"num_tokens": 8042823.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.4966999053955075,
|
|
"epoch": 3.669692385274836,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040942430973770323,
|
|
"loss": 5.3017,
|
|
"mean_token_accuracy": 0.16085462123155594,
|
|
"num_tokens": 8054516.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.421261405944824,
|
|
"epoch": 3.674735249621785,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004091060097554712,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.1668559268116951,
|
|
"num_tokens": 8064783.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.371801424026489,
|
|
"epoch": 3.679778113968734,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004087872929686673,
|
|
"loss": 5.1648,
|
|
"mean_token_accuracy": 0.16690610349178314,
|
|
"num_tokens": 8076085.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.4291423797607425,
|
|
"epoch": 3.6848209783156833,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040846816036787056,
|
|
"loss": 5.2886,
|
|
"mean_token_accuracy": 0.15664487779140474,
|
|
"num_tokens": 8088218.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.381323003768921,
|
|
"epoch": 3.6898638426626325,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004081486129449527,
|
|
"loss": 5.2096,
|
|
"mean_token_accuracy": 0.16191145330667495,
|
|
"num_tokens": 8099768.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.407671165466309,
|
|
"epoch": 3.6949067070095816,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040782865169307423,
|
|
"loss": 5.1982,
|
|
"mean_token_accuracy": 0.16725882291793823,
|
|
"num_tokens": 8109858.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.516709995269776,
|
|
"epoch": 3.6999495713565302,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004075082776066822,
|
|
"loss": 5.315,
|
|
"mean_token_accuracy": 0.16229137778282166,
|
|
"num_tokens": 8120262.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.50908842086792,
|
|
"epoch": 3.7049924357034794,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004071874916815068,
|
|
"loss": 5.249,
|
|
"mean_token_accuracy": 0.16730919629335403,
|
|
"num_tokens": 8131483.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.462291193008423,
|
|
"epoch": 3.7100353000504285,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040686629491455787,
|
|
"loss": 5.3096,
|
|
"mean_token_accuracy": 0.15786902457475663,
|
|
"num_tokens": 8143431.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.487177610397339,
|
|
"epoch": 3.7150781643973776,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040654468830412247,
|
|
"loss": 5.3054,
|
|
"mean_token_accuracy": 0.16073084771633148,
|
|
"num_tokens": 8154596.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.393495893478393,
|
|
"epoch": 3.7201210287443267,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00040622267284976134,
|
|
"loss": 5.285,
|
|
"mean_token_accuracy": 0.1602349177002907,
|
|
"num_tokens": 8165509.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.4206235885620115,
|
|
"epoch": 3.725163893091276,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040590024955230596,
|
|
"loss": 5.1794,
|
|
"mean_token_accuracy": 0.16264848709106444,
|
|
"num_tokens": 8176275.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.384602308273315,
|
|
"epoch": 3.730206757438225,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004055774194138554,
|
|
"loss": 5.2701,
|
|
"mean_token_accuracy": 0.1620113343000412,
|
|
"num_tokens": 8187820.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.361208868026734,
|
|
"epoch": 3.735249621785174,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004052541834377731,
|
|
"loss": 5.1528,
|
|
"mean_token_accuracy": 0.16657100170850753,
|
|
"num_tokens": 8197750.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.421066045761108,
|
|
"epoch": 3.740292486132123,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040493054262868397,
|
|
"loss": 5.2555,
|
|
"mean_token_accuracy": 0.16027550846338273,
|
|
"num_tokens": 8209275.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.420915985107422,
|
|
"epoch": 3.7453353504790723,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040460649799247116,
|
|
"loss": 5.3062,
|
|
"mean_token_accuracy": 0.15899969637393951,
|
|
"num_tokens": 8221280.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.433041524887085,
|
|
"epoch": 3.7503782148260214,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004042820505362728,
|
|
"loss": 5.1886,
|
|
"mean_token_accuracy": 0.16738417595624924,
|
|
"num_tokens": 8231679.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.381321430206299,
|
|
"epoch": 3.7554210791729705,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004039572012684792,
|
|
"loss": 5.1863,
|
|
"mean_token_accuracy": 0.16833597272634507,
|
|
"num_tokens": 8242014.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.396518754959106,
|
|
"epoch": 3.760463943519919,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040363195119872916,
|
|
"loss": 5.1676,
|
|
"mean_token_accuracy": 0.16874904185533524,
|
|
"num_tokens": 8253128.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.333758163452148,
|
|
"epoch": 3.7655068078668683,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040330630133790757,
|
|
"loss": 5.1154,
|
|
"mean_token_accuracy": 0.17616755962371827,
|
|
"num_tokens": 8264643.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.437575769424439,
|
|
"epoch": 3.7705496722138174,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040298025269814165,
|
|
"loss": 5.2636,
|
|
"mean_token_accuracy": 0.16203719973564149,
|
|
"num_tokens": 8275009.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.373467350006104,
|
|
"epoch": 3.7755925365607665,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004026538062927981,
|
|
"loss": 5.2119,
|
|
"mean_token_accuracy": 0.1589460164308548,
|
|
"num_tokens": 8286321.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.402272272109985,
|
|
"epoch": 3.7806354009077157,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040232696313647985,
|
|
"loss": 5.2194,
|
|
"mean_token_accuracy": 0.1603735625743866,
|
|
"num_tokens": 8297216.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.46411166191101,
|
|
"epoch": 3.7856782652546648,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 5.2181,
|
|
"mean_token_accuracy": 0.160633085668087,
|
|
"num_tokens": 8309053.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.400403308868408,
|
|
"epoch": 3.7907211296016134,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004016720906354937,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.16017768532037735,
|
|
"num_tokens": 8318840.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.415168762207031,
|
|
"epoch": 3.7957639939485626,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004013440633261845,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.16871240139007568,
|
|
"num_tokens": 8329127.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.409310340881348,
|
|
"epoch": 3.8008068582955117,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00040101564333661213,
|
|
"loss": 5.1986,
|
|
"mean_token_accuracy": 0.167653651535511,
|
|
"num_tokens": 8340651.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.35351710319519,
|
|
"epoch": 3.805849722642461,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040068683168751316,
|
|
"loss": 5.2088,
|
|
"mean_token_accuracy": 0.16815593391656875,
|
|
"num_tokens": 8351399.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.4036847114562985,
|
|
"epoch": 3.81089258698941,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040035762940084205,
|
|
"loss": 5.1883,
|
|
"mean_token_accuracy": 0.16491729170084,
|
|
"num_tokens": 8361756.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.446819829940796,
|
|
"epoch": 3.815935451336359,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000400028037499767,
|
|
"loss": 5.2239,
|
|
"mean_token_accuracy": 0.16567778438329697,
|
|
"num_tokens": 8372408.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.3853254318237305,
|
|
"epoch": 3.820978315683308,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003996980570086673,
|
|
"loss": 5.1576,
|
|
"mean_token_accuracy": 0.1717057928442955,
|
|
"num_tokens": 8382788.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.3489830493927,
|
|
"epoch": 3.8260211800302573,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039936768895312987,
|
|
"loss": 5.2317,
|
|
"mean_token_accuracy": 0.16352340281009675,
|
|
"num_tokens": 8393693.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.392658329010009,
|
|
"epoch": 3.8310640443772064,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039903693435994626,
|
|
"loss": 5.2109,
|
|
"mean_token_accuracy": 0.17214041948318481,
|
|
"num_tokens": 8405163.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.402809000015258,
|
|
"epoch": 3.8361069087241555,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00039870579425710937,
|
|
"loss": 5.1963,
|
|
"mean_token_accuracy": 0.16173276156187058,
|
|
"num_tokens": 8416987.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.363797426223755,
|
|
"epoch": 3.8411497730711046,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003983742696738103,
|
|
"loss": 5.2177,
|
|
"mean_token_accuracy": 0.1640700489282608,
|
|
"num_tokens": 8428864.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.416312217712402,
|
|
"epoch": 3.8461926374180537,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003980423616404351,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.16375678032636642,
|
|
"num_tokens": 8438970.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.49581208229065,
|
|
"epoch": 3.8512355017650024,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003977100711885616,
|
|
"loss": 5.304,
|
|
"mean_token_accuracy": 0.15686767250299455,
|
|
"num_tokens": 8448935.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.418037557601929,
|
|
"epoch": 3.8562783661119515,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039737739935095614,
|
|
"loss": 5.2484,
|
|
"mean_token_accuracy": 0.16554856598377227,
|
|
"num_tokens": 8458908.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.415468406677246,
|
|
"epoch": 3.8613212304589006,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003970443471615705,
|
|
"loss": 5.1838,
|
|
"mean_token_accuracy": 0.1734945371747017,
|
|
"num_tokens": 8469161.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.352152872085571,
|
|
"epoch": 3.8663640948058497,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003967109156555386,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.16743157356977462,
|
|
"num_tokens": 8479821.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.417977142333984,
|
|
"epoch": 3.871406959152799,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039637710586917324,
|
|
"loss": 5.2605,
|
|
"mean_token_accuracy": 0.16112958639860153,
|
|
"num_tokens": 8489837.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.440296983718872,
|
|
"epoch": 3.876449823499748,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00039604291883996287,
|
|
"loss": 5.2416,
|
|
"mean_token_accuracy": 0.16606980115175246,
|
|
"num_tokens": 8501241.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.442011117935181,
|
|
"epoch": 3.8814926878466967,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039570835560656854,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.16614895313978195,
|
|
"num_tokens": 8512558.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.3762805461883545,
|
|
"epoch": 3.8865355521936458,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003953734172088205,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.16708417236804962,
|
|
"num_tokens": 8522870.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.390462827682495,
|
|
"epoch": 3.891578416540595,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039503810468771494,
|
|
"loss": 5.2317,
|
|
"mean_token_accuracy": 0.16003523766994476,
|
|
"num_tokens": 8533749.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.454382276535034,
|
|
"epoch": 3.896621280887544,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003947024190854111,
|
|
"loss": 5.2399,
|
|
"mean_token_accuracy": 0.1711591064929962,
|
|
"num_tokens": 8544700.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.412538242340088,
|
|
"epoch": 3.901664145234493,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039436636144522737,
|
|
"loss": 5.2522,
|
|
"mean_token_accuracy": 0.15667204558849335,
|
|
"num_tokens": 8555236.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.39568943977356,
|
|
"epoch": 3.9067070095814422,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039402993281163875,
|
|
"loss": 5.2712,
|
|
"mean_token_accuracy": 0.1599903002381325,
|
|
"num_tokens": 8566597.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.431813669204712,
|
|
"epoch": 3.9117498739283914,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003936931342302732,
|
|
"loss": 5.2221,
|
|
"mean_token_accuracy": 0.15816849023103713,
|
|
"num_tokens": 8577297.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.474480056762696,
|
|
"epoch": 3.9167927382753405,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003933559667479084,
|
|
"loss": 5.2422,
|
|
"mean_token_accuracy": 0.16582781374454497,
|
|
"num_tokens": 8587145.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.359107446670532,
|
|
"epoch": 3.9218356026222896,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00039301843141246876,
|
|
"loss": 5.1215,
|
|
"mean_token_accuracy": 0.17119301855564117,
|
|
"num_tokens": 8597410.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.380753707885742,
|
|
"epoch": 3.9268784669692387,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039268052927302185,
|
|
"loss": 5.1872,
|
|
"mean_token_accuracy": 0.16905289590358735,
|
|
"num_tokens": 8607839.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.393512725830078,
|
|
"epoch": 3.931921331316188,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003923422613797753,
|
|
"loss": 5.1694,
|
|
"mean_token_accuracy": 0.16536030620336534,
|
|
"num_tokens": 8618517.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.32657642364502,
|
|
"epoch": 3.936964195663137,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003920036287840734,
|
|
"loss": 5.1453,
|
|
"mean_token_accuracy": 0.17355929166078568,
|
|
"num_tokens": 8629120.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.388219213485717,
|
|
"epoch": 3.9420070600100856,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039166463253839434,
|
|
"loss": 5.2074,
|
|
"mean_token_accuracy": 0.17029678523540498,
|
|
"num_tokens": 8640503.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.37358341217041,
|
|
"epoch": 3.9470499243570347,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003913252736963462,
|
|
"loss": 5.2029,
|
|
"mean_token_accuracy": 0.17363450974225997,
|
|
"num_tokens": 8652575.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.416703367233277,
|
|
"epoch": 3.952092788703984,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000390985553312664,
|
|
"loss": 5.1198,
|
|
"mean_token_accuracy": 0.16988059133291245,
|
|
"num_tokens": 8663646.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.38117995262146,
|
|
"epoch": 3.957135653050933,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003906454724432065,
|
|
"loss": 5.2058,
|
|
"mean_token_accuracy": 0.16442517936229706,
|
|
"num_tokens": 8673953.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.350530624389648,
|
|
"epoch": 3.962178517397882,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039030503214495296,
|
|
"loss": 5.214,
|
|
"mean_token_accuracy": 0.16484454721212388,
|
|
"num_tokens": 8685388.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.380281543731689,
|
|
"epoch": 3.967221381744831,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003899642334759996,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.16580954045057297,
|
|
"num_tokens": 8695205.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.3921104907989506,
|
|
"epoch": 3.97226424609178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003896230774955567,
|
|
"loss": 5.184,
|
|
"mean_token_accuracy": 0.16484033167362214,
|
|
"num_tokens": 8706118.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.371186542510986,
|
|
"epoch": 3.977307110438729,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038928156526394476,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.16910630613565444,
|
|
"num_tokens": 8716926.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.447496032714843,
|
|
"epoch": 3.982349974785678,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003889396978425918,
|
|
"loss": 5.2969,
|
|
"mean_token_accuracy": 0.16213430911302568,
|
|
"num_tokens": 8729409.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.4057355403900145,
|
|
"epoch": 3.987392839132627,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003885974762940295,
|
|
"loss": 5.235,
|
|
"mean_token_accuracy": 0.1684113085269928,
|
|
"num_tokens": 8740140.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.4019121646881105,
|
|
"epoch": 3.9924357034795763,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003882549016818905,
|
|
"loss": 5.1196,
|
|
"mean_token_accuracy": 0.16654850393533707,
|
|
"num_tokens": 8749176.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.355155277252197,
|
|
"epoch": 3.9974785678265254,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038791197507090456,
|
|
"loss": 5.2317,
|
|
"mean_token_accuracy": 0.17030496895313263,
|
|
"num_tokens": 8761575.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.3184362517462835,
|
|
"epoch": 4.00201714573878,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00038756869752689527,
|
|
"loss": 5.0291,
|
|
"mean_token_accuracy": 0.17253867619567448,
|
|
"num_tokens": 8771147.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.395121049880982,
|
|
"epoch": 4.007060010085729,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003872250701167775,
|
|
"loss": 4.8855,
|
|
"mean_token_accuracy": 0.18297589272260667,
|
|
"num_tokens": 8781936.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.361803436279297,
|
|
"epoch": 4.012102874432678,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038688109390855297,
|
|
"loss": 4.8943,
|
|
"mean_token_accuracy": 0.18275870084762574,
|
|
"num_tokens": 8791492.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.365276145935058,
|
|
"epoch": 4.017145738779627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003865367699713077,
|
|
"loss": 4.8808,
|
|
"mean_token_accuracy": 0.18234068602323533,
|
|
"num_tokens": 8802552.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.399391603469849,
|
|
"epoch": 4.022188603126576,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038619209937520836,
|
|
"loss": 4.9509,
|
|
"mean_token_accuracy": 0.18303546756505967,
|
|
"num_tokens": 8813885.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.3691075325012205,
|
|
"epoch": 4.027231467473525,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038584708319149924,
|
|
"loss": 4.911,
|
|
"mean_token_accuracy": 0.18313807249069214,
|
|
"num_tokens": 8825503.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.408253812789917,
|
|
"epoch": 4.032274331820474,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038550172249249856,
|
|
"loss": 4.9498,
|
|
"mean_token_accuracy": 0.17092643231153487,
|
|
"num_tokens": 8835581.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.032274331820474,
|
|
"eval_entropy": 5.275534075331748,
|
|
"eval_loss": 5.767982482910156,
|
|
"eval_mean_token_accuracy": 0.14656087610178625,
|
|
"eval_num_tokens": 8835581.0,
|
|
"eval_runtime": 2.632,
|
|
"eval_samples_per_second": 1192.261,
|
|
"eval_steps_per_second": 149.318,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9910,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.4749689224448e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|