bd27151216
Model: fpadovani/ita-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.5123550176500253,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 5510751532416000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|