20145 lines
550 KiB
JSON
20145 lines
550 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.940825459970164,
|
|
"eval_steps": 1000,
|
|
"global_step": 10000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742125511169434,
|
|
"epoch": 0.004972650422675286,
|
|
"grad_norm": 4.59375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7551,
|
|
"mean_token_accuracy": 0.00012019231216982007,
|
|
"num_tokens": 9311.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742149543762206,
|
|
"epoch": 0.009945300845350571,
|
|
"grad_norm": 3.75,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7919,
|
|
"mean_token_accuracy": 0.00014849193394184113,
|
|
"num_tokens": 20562.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742162990570069,
|
|
"epoch": 0.014917951268025857,
|
|
"grad_norm": 4.25,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.763,
|
|
"mean_token_accuracy": 0.0001310616033151746,
|
|
"num_tokens": 31490.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742100620269776,
|
|
"epoch": 0.019890601690701143,
|
|
"grad_norm": 4.8125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6827,
|
|
"mean_token_accuracy": 0.0001663893461227417,
|
|
"num_tokens": 40741.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742075634002685,
|
|
"epoch": 0.02486325211337643,
|
|
"grad_norm": 4.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.6327,
|
|
"mean_token_accuracy": 0.003865147603210062,
|
|
"num_tokens": 50749.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741719722747803,
|
|
"epoch": 0.029835902536051714,
|
|
"grad_norm": 3.8125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.5209,
|
|
"mean_token_accuracy": 0.03559233639389277,
|
|
"num_tokens": 60296.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740651512145996,
|
|
"epoch": 0.034808552958727,
|
|
"grad_norm": 3.15625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.4449,
|
|
"mean_token_accuracy": 0.043226226791739465,
|
|
"num_tokens": 70797.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.736448192596436,
|
|
"epoch": 0.039781203381402286,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.3251,
|
|
"mean_token_accuracy": 0.04369042068719864,
|
|
"num_tokens": 80815.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.726639556884766,
|
|
"epoch": 0.04475385380407757,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.2125,
|
|
"mean_token_accuracy": 0.04711066335439682,
|
|
"num_tokens": 90326.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.713320446014404,
|
|
"epoch": 0.04972650422675286,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.1807,
|
|
"mean_token_accuracy": 0.04810092821717262,
|
|
"num_tokens": 99800.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.700002956390382,
|
|
"epoch": 0.05469915464942814,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.1208,
|
|
"mean_token_accuracy": 0.04054830018430948,
|
|
"num_tokens": 110636.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.693144798278809,
|
|
"epoch": 0.05967180507210343,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.0616,
|
|
"mean_token_accuracy": 0.045356238074600694,
|
|
"num_tokens": 120269.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.688432979583741,
|
|
"epoch": 0.06464445549477872,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.032,
|
|
"mean_token_accuracy": 0.0448658213019371,
|
|
"num_tokens": 130176.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.681407642364501,
|
|
"epoch": 0.069617105917454,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.9591,
|
|
"mean_token_accuracy": 0.0455975791439414,
|
|
"num_tokens": 140020.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.67218999862671,
|
|
"epoch": 0.07458975634012929,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.9701,
|
|
"mean_token_accuracy": 0.04541345350444317,
|
|
"num_tokens": 150713.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.663935947418214,
|
|
"epoch": 0.07956240676280457,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.8558,
|
|
"mean_token_accuracy": 0.048045172542333606,
|
|
"num_tokens": 159828.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.650499153137208,
|
|
"epoch": 0.08453505718547986,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.811,
|
|
"mean_token_accuracy": 0.04568581245839596,
|
|
"num_tokens": 169909.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.636698341369629,
|
|
"epoch": 0.08950770760815514,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.7401,
|
|
"mean_token_accuracy": 0.04568461328744888,
|
|
"num_tokens": 179462.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.620390605926513,
|
|
"epoch": 0.09448035803083044,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.7037,
|
|
"mean_token_accuracy": 0.05530158616602421,
|
|
"num_tokens": 188965.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.597137260437012,
|
|
"epoch": 0.09945300845350571,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.7163,
|
|
"mean_token_accuracy": 0.054649893566966055,
|
|
"num_tokens": 200382.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.574395561218262,
|
|
"epoch": 0.10442565887618101,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.52,
|
|
"mean_token_accuracy": 0.06198187619447708,
|
|
"num_tokens": 209718.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.530473995208741,
|
|
"epoch": 0.10939830929885629,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.4448,
|
|
"mean_token_accuracy": 0.06471945084631443,
|
|
"num_tokens": 219216.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.469701290130615,
|
|
"epoch": 0.11437095972153158,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.3904,
|
|
"mean_token_accuracy": 0.061312200129032136,
|
|
"num_tokens": 228219.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.413739681243896,
|
|
"epoch": 0.11934361014420686,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.3035,
|
|
"mean_token_accuracy": 0.05876123383641243,
|
|
"num_tokens": 237776.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.371811962127685,
|
|
"epoch": 0.12431626056688215,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.2207,
|
|
"mean_token_accuracy": 0.06525652445852756,
|
|
"num_tokens": 248411.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.323804950714111,
|
|
"epoch": 0.12928891098955744,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.2051,
|
|
"mean_token_accuracy": 0.056814908236265185,
|
|
"num_tokens": 258797.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.282233047485352,
|
|
"epoch": 0.13426156141223272,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.0778,
|
|
"mean_token_accuracy": 0.05855576582252979,
|
|
"num_tokens": 269228.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.210856342315674,
|
|
"epoch": 0.139234211834908,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 9.012,
|
|
"mean_token_accuracy": 0.06329346187412739,
|
|
"num_tokens": 279191.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.10884141921997,
|
|
"epoch": 0.14420686225758328,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.902,
|
|
"mean_token_accuracy": 0.0638952475041151,
|
|
"num_tokens": 289382.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.034085178375244,
|
|
"epoch": 0.14917951268025859,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.8052,
|
|
"mean_token_accuracy": 0.06600089073181152,
|
|
"num_tokens": 299163.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.948583889007569,
|
|
"epoch": 0.15415216310293386,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.8105,
|
|
"mean_token_accuracy": 0.05745151527225971,
|
|
"num_tokens": 308551.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.804897499084472,
|
|
"epoch": 0.15912481352560914,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.6931,
|
|
"mean_token_accuracy": 0.058856929466128347,
|
|
"num_tokens": 318427.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.729992198944093,
|
|
"epoch": 0.16409746394828442,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.6865,
|
|
"mean_token_accuracy": 0.06036209538578987,
|
|
"num_tokens": 327539.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.571413707733154,
|
|
"epoch": 0.16907011437095973,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.5792,
|
|
"mean_token_accuracy": 0.06282380670309066,
|
|
"num_tokens": 336676.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.379988861083984,
|
|
"epoch": 0.174042764793635,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.5458,
|
|
"mean_token_accuracy": 0.07260882444679737,
|
|
"num_tokens": 345914.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.303316020965577,
|
|
"epoch": 0.1790154152163103,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.5171,
|
|
"mean_token_accuracy": 0.06382278949022294,
|
|
"num_tokens": 355982.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.262152862548827,
|
|
"epoch": 0.1839880656389856,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.4934,
|
|
"mean_token_accuracy": 0.05823831483721733,
|
|
"num_tokens": 365398.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.157662105560302,
|
|
"epoch": 0.18896071606166087,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.4637,
|
|
"mean_token_accuracy": 0.06616585291922092,
|
|
"num_tokens": 374519.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 9.057036304473877,
|
|
"epoch": 0.19393336648433615,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.4285,
|
|
"mean_token_accuracy": 0.06589658446609974,
|
|
"num_tokens": 383536.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.975014305114746,
|
|
"epoch": 0.19890601690701143,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.4664,
|
|
"mean_token_accuracy": 0.06985955312848091,
|
|
"num_tokens": 393703.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 9.013352203369141,
|
|
"epoch": 0.20387866732968674,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.484,
|
|
"mean_token_accuracy": 0.05608684942126274,
|
|
"num_tokens": 404065.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.898391628265381,
|
|
"epoch": 0.20885131775236201,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.3867,
|
|
"mean_token_accuracy": 0.06348490491509437,
|
|
"num_tokens": 413869.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.8826735496521,
|
|
"epoch": 0.2138239681750373,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.4475,
|
|
"mean_token_accuracy": 0.06403781101107597,
|
|
"num_tokens": 424251.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.847267246246338,
|
|
"epoch": 0.21879661859771257,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.3625,
|
|
"mean_token_accuracy": 0.06728239133954048,
|
|
"num_tokens": 432530.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.82140293121338,
|
|
"epoch": 0.22376926902038788,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.3576,
|
|
"mean_token_accuracy": 0.06472381576895714,
|
|
"num_tokens": 442335.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.79792947769165,
|
|
"epoch": 0.22874191944306316,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.3602,
|
|
"mean_token_accuracy": 0.06484326869249343,
|
|
"num_tokens": 452048.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.770851135253906,
|
|
"epoch": 0.23371456986573844,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.2701,
|
|
"mean_token_accuracy": 0.06962885931134224,
|
|
"num_tokens": 460840.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.781954383850097,
|
|
"epoch": 0.23868722028841372,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.3541,
|
|
"mean_token_accuracy": 0.07056505233049393,
|
|
"num_tokens": 470198.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.723297309875488,
|
|
"epoch": 0.24365987071108902,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.3298,
|
|
"mean_token_accuracy": 0.06994973197579384,
|
|
"num_tokens": 481030.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.75924997329712,
|
|
"epoch": 0.2486325211337643,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.2614,
|
|
"mean_token_accuracy": 0.07425586357712746,
|
|
"num_tokens": 490411.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.648718643188477,
|
|
"epoch": 0.2536051715564396,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.3018,
|
|
"mean_token_accuracy": 0.07575512938201427,
|
|
"num_tokens": 499551.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.756513404846192,
|
|
"epoch": 0.2585778219791149,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.3074,
|
|
"mean_token_accuracy": 0.06927388943731785,
|
|
"num_tokens": 508885.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.711030769348145,
|
|
"epoch": 0.26355047240179014,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.3146,
|
|
"mean_token_accuracy": 0.06607123203575611,
|
|
"num_tokens": 518291.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.714559173583984,
|
|
"epoch": 0.26852312282446544,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.2204,
|
|
"mean_token_accuracy": 0.07338580824434757,
|
|
"num_tokens": 528035.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.609884071350098,
|
|
"epoch": 0.27349577324714075,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.2632,
|
|
"mean_token_accuracy": 0.07376815192401409,
|
|
"num_tokens": 538410.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.679876708984375,
|
|
"epoch": 0.278468423669816,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.233,
|
|
"mean_token_accuracy": 0.07235329188406467,
|
|
"num_tokens": 549697.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.666294574737549,
|
|
"epoch": 0.2834410740924913,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.1295,
|
|
"mean_token_accuracy": 0.07369526959955693,
|
|
"num_tokens": 558932.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.43987340927124,
|
|
"epoch": 0.28841372451516656,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.2142,
|
|
"mean_token_accuracy": 0.07114675119519234,
|
|
"num_tokens": 568441.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.647719287872315,
|
|
"epoch": 0.29338637493784187,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.276,
|
|
"mean_token_accuracy": 0.07263872623443604,
|
|
"num_tokens": 577861.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.590884780883789,
|
|
"epoch": 0.29835902536051717,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.2563,
|
|
"mean_token_accuracy": 0.06960625350475311,
|
|
"num_tokens": 587858.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.584195327758788,
|
|
"epoch": 0.3033316757831924,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.2886,
|
|
"mean_token_accuracy": 0.07146217562258243,
|
|
"num_tokens": 597806.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.64070110321045,
|
|
"epoch": 0.30830432620586773,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.2035,
|
|
"mean_token_accuracy": 0.07901516482234001,
|
|
"num_tokens": 607408.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.609154510498048,
|
|
"epoch": 0.31327697662854304,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.1512,
|
|
"mean_token_accuracy": 0.07548471316695213,
|
|
"num_tokens": 617862.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.547056102752686,
|
|
"epoch": 0.3182496270512183,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.1592,
|
|
"mean_token_accuracy": 0.0737653348594904,
|
|
"num_tokens": 627128.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.555311012268067,
|
|
"epoch": 0.3232222774738936,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.1651,
|
|
"mean_token_accuracy": 0.0736689142882824,
|
|
"num_tokens": 636775.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.67299509048462,
|
|
"epoch": 0.32819492789656884,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.3453,
|
|
"mean_token_accuracy": 0.0731102954596281,
|
|
"num_tokens": 645906.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.630266380310058,
|
|
"epoch": 0.33316757831924415,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1417,
|
|
"mean_token_accuracy": 0.07412039376795292,
|
|
"num_tokens": 655601.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.552510166168213,
|
|
"epoch": 0.33814022874191946,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.1802,
|
|
"mean_token_accuracy": 0.07779353931546211,
|
|
"num_tokens": 665682.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.50169677734375,
|
|
"epoch": 0.3431128791645947,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.1144,
|
|
"mean_token_accuracy": 0.07909605242311954,
|
|
"num_tokens": 675200.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.562351989746094,
|
|
"epoch": 0.34808552958727,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.13,
|
|
"mean_token_accuracy": 0.07913120314478875,
|
|
"num_tokens": 685590.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.483703517913819,
|
|
"epoch": 0.3530581800099453,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.1169,
|
|
"mean_token_accuracy": 0.0802781730890274,
|
|
"num_tokens": 695665.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.588793182373047,
|
|
"epoch": 0.3580308304326206,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1612,
|
|
"mean_token_accuracy": 0.07845537327229976,
|
|
"num_tokens": 705850.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.599618053436279,
|
|
"epoch": 0.3630034808552959,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.1544,
|
|
"mean_token_accuracy": 0.0771234653890133,
|
|
"num_tokens": 716010.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.45508108139038,
|
|
"epoch": 0.3679761312779712,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.1025,
|
|
"mean_token_accuracy": 0.07987580895423889,
|
|
"num_tokens": 724908.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.501910209655762,
|
|
"epoch": 0.37294878170064644,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.0349,
|
|
"mean_token_accuracy": 0.08155370876193047,
|
|
"num_tokens": 733479.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.493549156188966,
|
|
"epoch": 0.37792143212332174,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.0843,
|
|
"mean_token_accuracy": 0.07720053121447563,
|
|
"num_tokens": 743365.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.546761417388916,
|
|
"epoch": 0.382894082545997,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.1308,
|
|
"mean_token_accuracy": 0.08084944300353528,
|
|
"num_tokens": 753720.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.440972423553466,
|
|
"epoch": 0.3878667329686723,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.0218,
|
|
"mean_token_accuracy": 0.0810876727104187,
|
|
"num_tokens": 762524.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.480606746673583,
|
|
"epoch": 0.3928393833913476,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.078,
|
|
"mean_token_accuracy": 0.08098804876208306,
|
|
"num_tokens": 772549.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.41929931640625,
|
|
"epoch": 0.39781203381402286,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.0447,
|
|
"mean_token_accuracy": 0.0820227913558483,
|
|
"num_tokens": 782304.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.439020824432372,
|
|
"epoch": 0.40278468423669817,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.1502,
|
|
"mean_token_accuracy": 0.0791780199855566,
|
|
"num_tokens": 792010.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.44935884475708,
|
|
"epoch": 0.40775733465937347,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.0124,
|
|
"mean_token_accuracy": 0.08555513694882393,
|
|
"num_tokens": 801667.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.391554164886475,
|
|
"epoch": 0.4127299850820487,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0786,
|
|
"mean_token_accuracy": 0.08500046245753765,
|
|
"num_tokens": 812385.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.365203952789306,
|
|
"epoch": 0.41770263550472403,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9896,
|
|
"mean_token_accuracy": 0.08640162125229836,
|
|
"num_tokens": 822177.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.447068977355958,
|
|
"epoch": 0.4226752859273993,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.0883,
|
|
"mean_token_accuracy": 0.0818880956619978,
|
|
"num_tokens": 832576.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.378250980377198,
|
|
"epoch": 0.4276479363500746,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.9749,
|
|
"mean_token_accuracy": 0.08843559697270394,
|
|
"num_tokens": 842290.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.320431423187255,
|
|
"epoch": 0.4326205867727499,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0072,
|
|
"mean_token_accuracy": 0.0826775886118412,
|
|
"num_tokens": 852500.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.388307189941406,
|
|
"epoch": 0.43759323719542514,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.9291,
|
|
"mean_token_accuracy": 0.08810395300388336,
|
|
"num_tokens": 861401.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.331867504119874,
|
|
"epoch": 0.44256588761810045,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.9524,
|
|
"mean_token_accuracy": 0.0882642038166523,
|
|
"num_tokens": 871523.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.389120292663574,
|
|
"epoch": 0.44753853804077576,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.0221,
|
|
"mean_token_accuracy": 0.07971839047968388,
|
|
"num_tokens": 881384.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.411963558197021,
|
|
"epoch": 0.452511188463451,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.0297,
|
|
"mean_token_accuracy": 0.07957848943769932,
|
|
"num_tokens": 890511.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.3255184173584,
|
|
"epoch": 0.4574838388861263,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.9838,
|
|
"mean_token_accuracy": 0.08396280780434609,
|
|
"num_tokens": 900064.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.366711044311524,
|
|
"epoch": 0.46245648930880157,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.0173,
|
|
"mean_token_accuracy": 0.08197253122925759,
|
|
"num_tokens": 910892.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.427517795562744,
|
|
"epoch": 0.4674291397314769,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.0223,
|
|
"mean_token_accuracy": 0.08293579518795013,
|
|
"num_tokens": 920566.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.250725555419923,
|
|
"epoch": 0.4724017901541522,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.9271,
|
|
"mean_token_accuracy": 0.08331855908036231,
|
|
"num_tokens": 930609.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.39367904663086,
|
|
"epoch": 0.47737444057682743,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.9568,
|
|
"mean_token_accuracy": 0.08248500525951385,
|
|
"num_tokens": 940383.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.419192504882812,
|
|
"epoch": 0.48234709099950274,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.9373,
|
|
"mean_token_accuracy": 0.08317923471331597,
|
|
"num_tokens": 951302.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.355427742004395,
|
|
"epoch": 0.48731974142217804,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.8796,
|
|
"mean_token_accuracy": 0.08742975443601608,
|
|
"num_tokens": 961126.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.324537467956542,
|
|
"epoch": 0.4922923918448533,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.8886,
|
|
"mean_token_accuracy": 0.08673487231135368,
|
|
"num_tokens": 970845.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.137747669219971,
|
|
"epoch": 0.4972650422675286,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.8409,
|
|
"mean_token_accuracy": 0.0927297443151474,
|
|
"num_tokens": 980919.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.347199821472168,
|
|
"epoch": 0.5022376926902039,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.8402,
|
|
"mean_token_accuracy": 0.09135852009057999,
|
|
"num_tokens": 990659.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.205520248413086,
|
|
"epoch": 0.5072103431128792,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.8968,
|
|
"mean_token_accuracy": 0.08830194622278213,
|
|
"num_tokens": 999540.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.256148338317871,
|
|
"epoch": 0.5121829935355544,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.8765,
|
|
"mean_token_accuracy": 0.08273696452379227,
|
|
"num_tokens": 1008688.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.282710742950439,
|
|
"epoch": 0.5171556439582298,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.7859,
|
|
"mean_token_accuracy": 0.09926471635699272,
|
|
"num_tokens": 1018310.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.081554937362672,
|
|
"epoch": 0.522128294380905,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.8937,
|
|
"mean_token_accuracy": 0.09027150124311448,
|
|
"num_tokens": 1027973.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.36844244003296,
|
|
"epoch": 0.5271009448035803,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.7986,
|
|
"mean_token_accuracy": 0.09017644748091698,
|
|
"num_tokens": 1038141.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.120935821533203,
|
|
"epoch": 0.5320735952262556,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.8933,
|
|
"mean_token_accuracy": 0.08534157127141953,
|
|
"num_tokens": 1047932.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.140996742248536,
|
|
"epoch": 0.5370462456489309,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.7198,
|
|
"mean_token_accuracy": 0.09635715857148171,
|
|
"num_tokens": 1056826.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.218596076965332,
|
|
"epoch": 0.5420188960716061,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.8424,
|
|
"mean_token_accuracy": 0.08895784765481948,
|
|
"num_tokens": 1066025.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.201892375946045,
|
|
"epoch": 0.5469915464942815,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.8624,
|
|
"mean_token_accuracy": 0.08816425576806068,
|
|
"num_tokens": 1075452.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.268182182312012,
|
|
"epoch": 0.5519641969169568,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.857,
|
|
"mean_token_accuracy": 0.08776547238230706,
|
|
"num_tokens": 1085300.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.11043586730957,
|
|
"epoch": 0.556936847339632,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.7534,
|
|
"mean_token_accuracy": 0.0959263451397419,
|
|
"num_tokens": 1094548.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.293241500854492,
|
|
"epoch": 0.5619094977623074,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.7717,
|
|
"mean_token_accuracy": 0.09139582961797714,
|
|
"num_tokens": 1104321.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.196415042877197,
|
|
"epoch": 0.5668821481849826,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8774,
|
|
"mean_token_accuracy": 0.08789880424737931,
|
|
"num_tokens": 1115047.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.211655712127685,
|
|
"epoch": 0.5718547986076579,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.835,
|
|
"mean_token_accuracy": 0.08990250006318093,
|
|
"num_tokens": 1124757.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.155872249603272,
|
|
"epoch": 0.5768274490303331,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.8204,
|
|
"mean_token_accuracy": 0.08568638414144517,
|
|
"num_tokens": 1134255.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.312313795089722,
|
|
"epoch": 0.5818000994530085,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.8616,
|
|
"mean_token_accuracy": 0.08821930140256881,
|
|
"num_tokens": 1143117.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.11629252433777,
|
|
"epoch": 0.5867727498756837,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.7429,
|
|
"mean_token_accuracy": 0.09225220009684562,
|
|
"num_tokens": 1153203.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.176023244857788,
|
|
"epoch": 0.591745400298359,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.7456,
|
|
"mean_token_accuracy": 0.09358465671539307,
|
|
"num_tokens": 1162839.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.032641792297364,
|
|
"epoch": 0.5967180507210343,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.7044,
|
|
"mean_token_accuracy": 0.09547256380319595,
|
|
"num_tokens": 1172580.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.247534561157227,
|
|
"epoch": 0.6016907011437096,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.6824,
|
|
"mean_token_accuracy": 0.09632320255041123,
|
|
"num_tokens": 1183715.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.140983057022094,
|
|
"epoch": 0.6066633515663848,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.8615,
|
|
"mean_token_accuracy": 0.08423814848065377,
|
|
"num_tokens": 1194217.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.09877405166626,
|
|
"epoch": 0.6116360019890602,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.7333,
|
|
"mean_token_accuracy": 0.09666020646691323,
|
|
"num_tokens": 1203297.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.166332054138184,
|
|
"epoch": 0.6166086524117355,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.8199,
|
|
"mean_token_accuracy": 0.08708633854985237,
|
|
"num_tokens": 1213394.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.102058935165406,
|
|
"epoch": 0.6215813028344107,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.7786,
|
|
"mean_token_accuracy": 0.09216831102967263,
|
|
"num_tokens": 1222771.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.104645824432373,
|
|
"epoch": 0.6265539532570861,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.7574,
|
|
"mean_token_accuracy": 0.08986867368221282,
|
|
"num_tokens": 1232171.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.156447982788086,
|
|
"epoch": 0.6315266036797613,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.6848,
|
|
"mean_token_accuracy": 0.09335485622286796,
|
|
"num_tokens": 1241640.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.9763720512390135,
|
|
"epoch": 0.6364992541024366,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.7093,
|
|
"mean_token_accuracy": 0.09285103902220726,
|
|
"num_tokens": 1251785.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.098073244094849,
|
|
"epoch": 0.6414719045251119,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.6956,
|
|
"mean_token_accuracy": 0.09296740368008613,
|
|
"num_tokens": 1262327.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.082415437698364,
|
|
"epoch": 0.6464445549477872,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.6441,
|
|
"mean_token_accuracy": 0.09833036586642266,
|
|
"num_tokens": 1270871.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.948732328414917,
|
|
"epoch": 0.6514172053704624,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.6771,
|
|
"mean_token_accuracy": 0.09403934106230735,
|
|
"num_tokens": 1281563.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.079524755477905,
|
|
"epoch": 0.6563898557931377,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.6798,
|
|
"mean_token_accuracy": 0.09144917652010917,
|
|
"num_tokens": 1290867.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.0707576751709,
|
|
"epoch": 0.661362506215813,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.6997,
|
|
"mean_token_accuracy": 0.09203669428825378,
|
|
"num_tokens": 1300384.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.101850461959838,
|
|
"epoch": 0.6663351566384883,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.6181,
|
|
"mean_token_accuracy": 0.09505762532353401,
|
|
"num_tokens": 1310443.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 8.008625078201295,
|
|
"epoch": 0.6713078070611636,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.6019,
|
|
"mean_token_accuracy": 0.10043176636099815,
|
|
"num_tokens": 1320345.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.995106840133667,
|
|
"epoch": 0.6762804574838389,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.748,
|
|
"mean_token_accuracy": 0.09445218369364738,
|
|
"num_tokens": 1330868.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 8.019366979598999,
|
|
"epoch": 0.6812531079065142,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6414,
|
|
"mean_token_accuracy": 0.09418305829167366,
|
|
"num_tokens": 1340605.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.03638129234314,
|
|
"epoch": 0.6862257583291894,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.6851,
|
|
"mean_token_accuracy": 0.09217890501022338,
|
|
"num_tokens": 1350248.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.857495450973511,
|
|
"epoch": 0.6911984087518648,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.6221,
|
|
"mean_token_accuracy": 0.09679363071918487,
|
|
"num_tokens": 1360251.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 8.019428586959839,
|
|
"epoch": 0.69617105917454,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.5909,
|
|
"mean_token_accuracy": 0.10328384339809418,
|
|
"num_tokens": 1370156.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.029846382141113,
|
|
"epoch": 0.7011437095972153,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7614,
|
|
"mean_token_accuracy": 0.08630104809999466,
|
|
"num_tokens": 1380346.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.969434547424316,
|
|
"epoch": 0.7061163600198906,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.6211,
|
|
"mean_token_accuracy": 0.09508981257677078,
|
|
"num_tokens": 1390958.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 8.002356576919556,
|
|
"epoch": 0.7110890104425659,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6858,
|
|
"mean_token_accuracy": 0.09678798392415047,
|
|
"num_tokens": 1400324.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 8.056915044784546,
|
|
"epoch": 0.7160616608652411,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.6014,
|
|
"mean_token_accuracy": 0.1017332375049591,
|
|
"num_tokens": 1410272.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.83396201133728,
|
|
"epoch": 0.7210343112879165,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.6202,
|
|
"mean_token_accuracy": 0.09909708872437477,
|
|
"num_tokens": 1420061.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.946973037719727,
|
|
"epoch": 0.7260069617105918,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.5531,
|
|
"mean_token_accuracy": 0.100252715498209,
|
|
"num_tokens": 1428771.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.849048757553101,
|
|
"epoch": 0.730979612133267,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.5472,
|
|
"mean_token_accuracy": 0.10210810899734497,
|
|
"num_tokens": 1438154.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.898545551300049,
|
|
"epoch": 0.7359522625559424,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.5355,
|
|
"mean_token_accuracy": 0.10233396664261818,
|
|
"num_tokens": 1447403.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 8.004046964645386,
|
|
"epoch": 0.7409249129786176,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6362,
|
|
"mean_token_accuracy": 0.09382892400026321,
|
|
"num_tokens": 1457443.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.92907829284668,
|
|
"epoch": 0.7458975634012929,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6128,
|
|
"mean_token_accuracy": 0.09921764805912972,
|
|
"num_tokens": 1466783.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 8.000693035125732,
|
|
"epoch": 0.7508702138239681,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.6289,
|
|
"mean_token_accuracy": 0.0994330421090126,
|
|
"num_tokens": 1476320.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.883225584030152,
|
|
"epoch": 0.7558428642466435,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.5361,
|
|
"mean_token_accuracy": 0.09907385855913162,
|
|
"num_tokens": 1487581.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.932295846939087,
|
|
"epoch": 0.7608155146693187,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6555,
|
|
"mean_token_accuracy": 0.09297903254628181,
|
|
"num_tokens": 1496455.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.92517466545105,
|
|
"epoch": 0.765788165091994,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.5886,
|
|
"mean_token_accuracy": 0.10096605643630027,
|
|
"num_tokens": 1506422.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.9216045379638675,
|
|
"epoch": 0.7707608155146694,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.5552,
|
|
"mean_token_accuracy": 0.09964245483279228,
|
|
"num_tokens": 1516593.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.880478572845459,
|
|
"epoch": 0.7757334659373446,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.6035,
|
|
"mean_token_accuracy": 0.09887727648019791,
|
|
"num_tokens": 1526011.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.903725481033325,
|
|
"epoch": 0.7807061163600199,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.5237,
|
|
"mean_token_accuracy": 0.0997832901775837,
|
|
"num_tokens": 1536018.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.783191680908203,
|
|
"epoch": 0.7856787667826952,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.4656,
|
|
"mean_token_accuracy": 0.10259042382240295,
|
|
"num_tokens": 1545568.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.89179482460022,
|
|
"epoch": 0.7906514172053705,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.4412,
|
|
"mean_token_accuracy": 0.10150249972939492,
|
|
"num_tokens": 1556389.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 8.001115608215333,
|
|
"epoch": 0.7956240676280457,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.5662,
|
|
"mean_token_accuracy": 0.09821114912629128,
|
|
"num_tokens": 1565587.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.793225812911987,
|
|
"epoch": 0.8005967180507211,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.4735,
|
|
"mean_token_accuracy": 0.1033795066177845,
|
|
"num_tokens": 1574911.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.673290205001831,
|
|
"epoch": 0.8055693684733963,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.4392,
|
|
"mean_token_accuracy": 0.1068122886121273,
|
|
"num_tokens": 1584477.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.83552131652832,
|
|
"epoch": 0.8105420188960716,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.5107,
|
|
"mean_token_accuracy": 0.10547451376914978,
|
|
"num_tokens": 1594408.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.824960470199585,
|
|
"epoch": 0.8155146693187469,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.4562,
|
|
"mean_token_accuracy": 0.10576815977692604,
|
|
"num_tokens": 1603494.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.820860242843628,
|
|
"epoch": 0.8204873197414222,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.5726,
|
|
"mean_token_accuracy": 0.09770831763744355,
|
|
"num_tokens": 1612994.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.901879072189331,
|
|
"epoch": 0.8254599701640974,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.4492,
|
|
"mean_token_accuracy": 0.10031033977866173,
|
|
"num_tokens": 1622150.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.778690671920776,
|
|
"epoch": 0.8304326205867727,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.4541,
|
|
"mean_token_accuracy": 0.10536026582121849,
|
|
"num_tokens": 1631040.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.7920936107635494,
|
|
"epoch": 0.8354052710094481,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.4141,
|
|
"mean_token_accuracy": 0.10774432495236397,
|
|
"num_tokens": 1641519.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.815248346328735,
|
|
"epoch": 0.8403779214321233,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.4791,
|
|
"mean_token_accuracy": 0.10060660392045975,
|
|
"num_tokens": 1652218.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.911854314804077,
|
|
"epoch": 0.8453505718547986,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.5812,
|
|
"mean_token_accuracy": 0.09471483305096626,
|
|
"num_tokens": 1663763.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.802376842498779,
|
|
"epoch": 0.8503232222774739,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.4903,
|
|
"mean_token_accuracy": 0.09759825989603996,
|
|
"num_tokens": 1674160.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.8265456199646,
|
|
"epoch": 0.8552958727001492,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5907,
|
|
"mean_token_accuracy": 0.10727964118123054,
|
|
"num_tokens": 1685138.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.904120016098022,
|
|
"epoch": 0.8602685231228244,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.4984,
|
|
"mean_token_accuracy": 0.09586141631007195,
|
|
"num_tokens": 1694847.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.810749959945679,
|
|
"epoch": 0.8652411735454998,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.4868,
|
|
"mean_token_accuracy": 0.10119377225637435,
|
|
"num_tokens": 1704503.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.715060997009277,
|
|
"epoch": 0.870213823968175,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.4403,
|
|
"mean_token_accuracy": 0.10172678083181382,
|
|
"num_tokens": 1714011.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.8425530910491945,
|
|
"epoch": 0.8751864743908503,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.4038,
|
|
"mean_token_accuracy": 0.1016093946993351,
|
|
"num_tokens": 1724254.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.704598665237427,
|
|
"epoch": 0.8801591248135257,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.4663,
|
|
"mean_token_accuracy": 0.09788005203008651,
|
|
"num_tokens": 1734696.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.717826509475708,
|
|
"epoch": 0.8851317752362009,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.4144,
|
|
"mean_token_accuracy": 0.10838304609060287,
|
|
"num_tokens": 1743650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.77191128730774,
|
|
"epoch": 0.8901044256588762,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.3787,
|
|
"mean_token_accuracy": 0.10693695098161697,
|
|
"num_tokens": 1753449.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.7488702774047855,
|
|
"epoch": 0.8950770760815515,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.5275,
|
|
"mean_token_accuracy": 0.10407080948352813,
|
|
"num_tokens": 1764104.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.685209608078003,
|
|
"epoch": 0.9000497265042268,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.2881,
|
|
"mean_token_accuracy": 0.11140932440757752,
|
|
"num_tokens": 1773172.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.809676027297973,
|
|
"epoch": 0.905022376926902,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.3725,
|
|
"mean_token_accuracy": 0.10514748990535736,
|
|
"num_tokens": 1783268.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.619670343399048,
|
|
"epoch": 0.9099950273495774,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.4064,
|
|
"mean_token_accuracy": 0.10027360618114471,
|
|
"num_tokens": 1792169.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.830101394653321,
|
|
"epoch": 0.9149676777722526,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4081,
|
|
"mean_token_accuracy": 0.10452633500099182,
|
|
"num_tokens": 1802280.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.709928941726685,
|
|
"epoch": 0.9199403281949279,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.4022,
|
|
"mean_token_accuracy": 0.10338718146085739,
|
|
"num_tokens": 1812086.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.6455628871917725,
|
|
"epoch": 0.9249129786176031,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.2895,
|
|
"mean_token_accuracy": 0.10942201912403107,
|
|
"num_tokens": 1821554.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.711309957504272,
|
|
"epoch": 0.9298856290402785,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4166,
|
|
"mean_token_accuracy": 0.10363708510994911,
|
|
"num_tokens": 1831088.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.681421422958374,
|
|
"epoch": 0.9348582794629537,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.4026,
|
|
"mean_token_accuracy": 0.11080866083502769,
|
|
"num_tokens": 1840606.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.614782381057739,
|
|
"epoch": 0.939830929885629,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.329,
|
|
"mean_token_accuracy": 0.11253917217254639,
|
|
"num_tokens": 1850343.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.6142534732818605,
|
|
"epoch": 0.9448035803083044,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.4489,
|
|
"mean_token_accuracy": 0.1043398305773735,
|
|
"num_tokens": 1860727.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.6588757038116455,
|
|
"epoch": 0.9497762307309796,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.3583,
|
|
"mean_token_accuracy": 0.10287954434752464,
|
|
"num_tokens": 1870703.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.660784912109375,
|
|
"epoch": 0.9547488811536549,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.3693,
|
|
"mean_token_accuracy": 0.10641058757901192,
|
|
"num_tokens": 1880694.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.6497437953948975,
|
|
"epoch": 0.9597215315763302,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.326,
|
|
"mean_token_accuracy": 0.10826451629400254,
|
|
"num_tokens": 1890195.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.647899389266968,
|
|
"epoch": 0.9646941819990055,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.3005,
|
|
"mean_token_accuracy": 0.10685399621725082,
|
|
"num_tokens": 1900221.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.509781742095948,
|
|
"epoch": 0.9696668324216807,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.3515,
|
|
"mean_token_accuracy": 0.10846741870045662,
|
|
"num_tokens": 1910122.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.747236728668213,
|
|
"epoch": 0.9746394828443561,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.3425,
|
|
"mean_token_accuracy": 0.10762734711170197,
|
|
"num_tokens": 1918814.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.556013917922973,
|
|
"epoch": 0.9796121332670313,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.3218,
|
|
"mean_token_accuracy": 0.11422898769378662,
|
|
"num_tokens": 1928446.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.570505523681641,
|
|
"epoch": 0.9845847836897066,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.3038,
|
|
"mean_token_accuracy": 0.1097981259226799,
|
|
"num_tokens": 1939060.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.66882095336914,
|
|
"epoch": 0.989557434112382,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.3081,
|
|
"mean_token_accuracy": 0.11472506150603294,
|
|
"num_tokens": 1948809.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.571680974960327,
|
|
"epoch": 0.9945300845350572,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.319,
|
|
"mean_token_accuracy": 0.1070831723511219,
|
|
"num_tokens": 1959232.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9945300845350572,
|
|
"eval_entropy": 7.6007331834960095,
|
|
"eval_loss": 7.42181396484375,
|
|
"eval_mean_token_accuracy": 0.10093975564988528,
|
|
"eval_num_tokens": 1959232.0,
|
|
"eval_runtime": 1.9993,
|
|
"eval_samples_per_second": 1733.12,
|
|
"eval_steps_per_second": 217.078,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.658403396606445,
|
|
"epoch": 0.9995027349577325,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999997830922735,
|
|
"loss": 7.25,
|
|
"mean_token_accuracy": 0.11331787556409836,
|
|
"num_tokens": 1969117.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.613470130496555,
|
|
"epoch": 1.0039781203381402,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999989019053515,
|
|
"loss": 7.2956,
|
|
"mean_token_accuracy": 0.10804594887627496,
|
|
"num_tokens": 1978202.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.4668820858001705,
|
|
"epoch": 1.0089507707608154,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999973428851536,
|
|
"loss": 7.0655,
|
|
"mean_token_accuracy": 0.12080970034003258,
|
|
"num_tokens": 1988085.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.552359199523925,
|
|
"epoch": 1.0139234211834909,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999951060363766,
|
|
"loss": 7.0751,
|
|
"mean_token_accuracy": 0.11799541562795639,
|
|
"num_tokens": 1996977.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.643637704849243,
|
|
"epoch": 1.0188960716061661,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999921913657592,
|
|
"loss": 7.1691,
|
|
"mean_token_accuracy": 0.11510241255164147,
|
|
"num_tokens": 2006622.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.610537433624268,
|
|
"epoch": 1.0238687220288414,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999885988820821,
|
|
"loss": 7.2389,
|
|
"mean_token_accuracy": 0.10962579101324081,
|
|
"num_tokens": 2016390.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.5456586360931395,
|
|
"epoch": 1.0288413724515166,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999843285961683,
|
|
"loss": 7.0696,
|
|
"mean_token_accuracy": 0.11251959577202797,
|
|
"num_tokens": 2026439.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.534916591644287,
|
|
"epoch": 1.0338140228741919,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999793805208822,
|
|
"loss": 7.1602,
|
|
"mean_token_accuracy": 0.12060272395610809,
|
|
"num_tokens": 2035611.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.490444660186768,
|
|
"epoch": 1.0387866732968671,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999737546711305,
|
|
"loss": 7.1803,
|
|
"mean_token_accuracy": 0.11013671979308129,
|
|
"num_tokens": 2047166.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.580088806152344,
|
|
"epoch": 1.0437593237195426,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999674510638618,
|
|
"loss": 7.0807,
|
|
"mean_token_accuracy": 0.11178387701511383,
|
|
"num_tokens": 2057216.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.479880428314209,
|
|
"epoch": 1.0487319741422179,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999604697180661,
|
|
"loss": 7.1495,
|
|
"mean_token_accuracy": 0.11265035048127174,
|
|
"num_tokens": 2067339.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.480796957015992,
|
|
"epoch": 1.053704624564893,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999528106547759,
|
|
"loss": 7.0348,
|
|
"mean_token_accuracy": 0.12159939408302307,
|
|
"num_tokens": 2077815.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.460572099685669,
|
|
"epoch": 1.0586772749875684,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999444738970644,
|
|
"loss": 7.143,
|
|
"mean_token_accuracy": 0.11123075261712075,
|
|
"num_tokens": 2087738.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.472248983383179,
|
|
"epoch": 1.0636499254102436,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999354594700474,
|
|
"loss": 7.01,
|
|
"mean_token_accuracy": 0.12233682870864868,
|
|
"num_tokens": 2097246.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.386615180969239,
|
|
"epoch": 1.0686225758329189,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999257674008817,
|
|
"loss": 6.967,
|
|
"mean_token_accuracy": 0.12300039678812028,
|
|
"num_tokens": 2106415.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.36377100944519,
|
|
"epoch": 1.0735952262555943,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999153977187656,
|
|
"loss": 6.965,
|
|
"mean_token_accuracy": 0.1262350156903267,
|
|
"num_tokens": 2115950.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.490091276168823,
|
|
"epoch": 1.0785678766782696,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499904350454939,
|
|
"loss": 7.1711,
|
|
"mean_token_accuracy": 0.11378273889422416,
|
|
"num_tokens": 2127896.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.436322164535523,
|
|
"epoch": 1.0835405271009448,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998926256426829,
|
|
"loss": 7.1047,
|
|
"mean_token_accuracy": 0.11285332590341568,
|
|
"num_tokens": 2137238.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.502145528793335,
|
|
"epoch": 1.08851317752362,
|
|
"grad_norm": 3.765625,
|
|
"learning_rate": 0.0004998802233173196,
|
|
"loss": 7.0764,
|
|
"mean_token_accuracy": 0.11752182170748711,
|
|
"num_tokens": 2147107.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.536004829406738,
|
|
"epoch": 1.0934858279462953,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998671435162123,
|
|
"loss": 7.138,
|
|
"mean_token_accuracy": 0.11741597354412078,
|
|
"num_tokens": 2157128.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.329599237442016,
|
|
"epoch": 1.0984584783689706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998533862787657,
|
|
"loss": 7.039,
|
|
"mean_token_accuracy": 0.11047931388020515,
|
|
"num_tokens": 2167981.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.4506263256073,
|
|
"epoch": 1.1034311287916458,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998389516464244,
|
|
"loss": 7.035,
|
|
"mean_token_accuracy": 0.11767932400107384,
|
|
"num_tokens": 2177659.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.400339078903198,
|
|
"epoch": 1.1084037792143213,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998238396626746,
|
|
"loss": 7.0515,
|
|
"mean_token_accuracy": 0.1187308482825756,
|
|
"num_tokens": 2186936.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.508743858337402,
|
|
"epoch": 1.1133764296369966,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998080503730427,
|
|
"loss": 7.0691,
|
|
"mean_token_accuracy": 0.12117458805441857,
|
|
"num_tokens": 2197126.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.3519598007202145,
|
|
"epoch": 1.1183490800596718,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997915838250956,
|
|
"loss": 7.1017,
|
|
"mean_token_accuracy": 0.1179386131465435,
|
|
"num_tokens": 2206337.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.337008666992188,
|
|
"epoch": 1.123321730482347,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997744400684406,
|
|
"loss": 7.0599,
|
|
"mean_token_accuracy": 0.11484007313847541,
|
|
"num_tokens": 2217112.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.483946228027344,
|
|
"epoch": 1.1282943809050223,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997566191547251,
|
|
"loss": 7.14,
|
|
"mean_token_accuracy": 0.11014441773295403,
|
|
"num_tokens": 2226239.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.518512201309204,
|
|
"epoch": 1.1332670313276976,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997381211376364,
|
|
"loss": 7.1411,
|
|
"mean_token_accuracy": 0.1147849939763546,
|
|
"num_tokens": 2235702.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.423879098892212,
|
|
"epoch": 1.138239681750373,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997189460729019,
|
|
"loss": 7.1075,
|
|
"mean_token_accuracy": 0.11887805461883545,
|
|
"num_tokens": 2246005.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.362666368484497,
|
|
"epoch": 1.1432123321730483,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996990940182884,
|
|
"loss": 7.0629,
|
|
"mean_token_accuracy": 0.11807055026292801,
|
|
"num_tokens": 2255534.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.444998407363892,
|
|
"epoch": 1.1481849825957235,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996785650336024,
|
|
"loss": 7.1205,
|
|
"mean_token_accuracy": 0.11631304249167443,
|
|
"num_tokens": 2266374.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.346836853027344,
|
|
"epoch": 1.1531576330183988,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996573591806898,
|
|
"loss": 6.9466,
|
|
"mean_token_accuracy": 0.12400897368788719,
|
|
"num_tokens": 2275614.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.340651130676269,
|
|
"epoch": 1.158130283441074,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996354765234353,
|
|
"loss": 7.0227,
|
|
"mean_token_accuracy": 0.12096358612179756,
|
|
"num_tokens": 2284403.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.371814632415772,
|
|
"epoch": 1.1631029338637493,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004996129171277628,
|
|
"loss": 7.0759,
|
|
"mean_token_accuracy": 0.11879568845033646,
|
|
"num_tokens": 2294153.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.376428079605103,
|
|
"epoch": 1.1680755842864246,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004995896810616349,
|
|
"loss": 7.0541,
|
|
"mean_token_accuracy": 0.11964974701404571,
|
|
"num_tokens": 2302897.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.507727003097534,
|
|
"epoch": 1.1730482347091,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995657683950528,
|
|
"loss": 7.1603,
|
|
"mean_token_accuracy": 0.11577019020915032,
|
|
"num_tokens": 2312483.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.195210409164429,
|
|
"epoch": 1.1780208851317753,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995411792000559,
|
|
"loss": 6.9033,
|
|
"mean_token_accuracy": 0.12615498080849646,
|
|
"num_tokens": 2321911.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.4335784912109375,
|
|
"epoch": 1.1829935355544505,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995159135507218,
|
|
"loss": 7.1605,
|
|
"mean_token_accuracy": 0.11667428612709045,
|
|
"num_tokens": 2331932.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.231681489944458,
|
|
"epoch": 1.1879661859771258,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499489971523166,
|
|
"loss": 6.9413,
|
|
"mean_token_accuracy": 0.12686287686228753,
|
|
"num_tokens": 2342264.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.425218868255615,
|
|
"epoch": 1.192938836399801,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004994633531955415,
|
|
"loss": 7.1048,
|
|
"mean_token_accuracy": 0.12006610035896301,
|
|
"num_tokens": 2353068.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.365460157394409,
|
|
"epoch": 1.1979114868224765,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994360586480391,
|
|
"loss": 6.9097,
|
|
"mean_token_accuracy": 0.12739017456769944,
|
|
"num_tokens": 2361765.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.33786907196045,
|
|
"epoch": 1.2028841372451518,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994080879628864,
|
|
"loss": 6.9982,
|
|
"mean_token_accuracy": 0.11641739755868911,
|
|
"num_tokens": 2372721.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.37421989440918,
|
|
"epoch": 1.207856787667827,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004993794412243481,
|
|
"loss": 7.0458,
|
|
"mean_token_accuracy": 0.12472002431750298,
|
|
"num_tokens": 2382288.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.310274744033814,
|
|
"epoch": 1.2128294380905023,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004993501185187255,
|
|
"loss": 7.0806,
|
|
"mean_token_accuracy": 0.11929435580968857,
|
|
"num_tokens": 2391953.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.415496921539306,
|
|
"epoch": 1.2178020885131775,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004993201199343565,
|
|
"loss": 7.1036,
|
|
"mean_token_accuracy": 0.11285114958882332,
|
|
"num_tokens": 2402705.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.298254489898682,
|
|
"epoch": 1.2227747389358528,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004992894455616149,
|
|
"loss": 7.0485,
|
|
"mean_token_accuracy": 0.11352694630622864,
|
|
"num_tokens": 2412654.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.451665878295898,
|
|
"epoch": 1.227747389358528,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004992580954929108,
|
|
"loss": 7.1418,
|
|
"mean_token_accuracy": 0.12201308831572533,
|
|
"num_tokens": 2422958.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.231288146972656,
|
|
"epoch": 1.2327200397812033,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004992260698226894,
|
|
"loss": 6.9725,
|
|
"mean_token_accuracy": 0.12452163249254226,
|
|
"num_tokens": 2432274.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.351004314422608,
|
|
"epoch": 1.2376926902038787,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991933686474315,
|
|
"loss": 7.1448,
|
|
"mean_token_accuracy": 0.11202256754040718,
|
|
"num_tokens": 2443818.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.367594051361084,
|
|
"epoch": 1.242665340626554,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004991599920656529,
|
|
"loss": 7.0227,
|
|
"mean_token_accuracy": 0.12140150368213654,
|
|
"num_tokens": 2453474.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.252175903320312,
|
|
"epoch": 1.2476379910492292,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004991259401779042,
|
|
"loss": 6.9866,
|
|
"mean_token_accuracy": 0.12221012189984322,
|
|
"num_tokens": 2462715.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.248402404785156,
|
|
"epoch": 1.2526106414719045,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0004990912130867704,
|
|
"loss": 6.9013,
|
|
"mean_token_accuracy": 0.12301692664623261,
|
|
"num_tokens": 2473256.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.315486764907837,
|
|
"epoch": 1.2575832918945797,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004990558108968703,
|
|
"loss": 6.9516,
|
|
"mean_token_accuracy": 0.11457495167851448,
|
|
"num_tokens": 2482475.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.3424886703491214,
|
|
"epoch": 1.2625559423172552,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004990197337148571,
|
|
"loss": 7.0004,
|
|
"mean_token_accuracy": 0.11988590955734253,
|
|
"num_tokens": 2493163.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.328835725784302,
|
|
"epoch": 1.2675285927399305,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000498982981649417,
|
|
"loss": 7.0519,
|
|
"mean_token_accuracy": 0.11444635018706321,
|
|
"num_tokens": 2502062.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.337628793716431,
|
|
"epoch": 1.2725012431626057,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004989455548112695,
|
|
"loss": 7.0278,
|
|
"mean_token_accuracy": 0.11994432136416436,
|
|
"num_tokens": 2511196.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.3562804698944095,
|
|
"epoch": 1.277473893585281,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004989074533131671,
|
|
"loss": 6.9628,
|
|
"mean_token_accuracy": 0.11777467504143715,
|
|
"num_tokens": 2520492.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.200314855575561,
|
|
"epoch": 1.2824465440079562,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004988686772698945,
|
|
"loss": 6.9752,
|
|
"mean_token_accuracy": 0.12104378938674927,
|
|
"num_tokens": 2530368.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.377554559707642,
|
|
"epoch": 1.2874191944306315,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004988292267982689,
|
|
"loss": 7.017,
|
|
"mean_token_accuracy": 0.11877000480890273,
|
|
"num_tokens": 2540734.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.213075399398804,
|
|
"epoch": 1.2923918448533067,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004987891020171387,
|
|
"loss": 7.0235,
|
|
"mean_token_accuracy": 0.11690100580453873,
|
|
"num_tokens": 2550907.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.401421070098877,
|
|
"epoch": 1.297364495275982,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004987483030473845,
|
|
"loss": 7.0648,
|
|
"mean_token_accuracy": 0.11762887313961982,
|
|
"num_tokens": 2560062.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.29442982673645,
|
|
"epoch": 1.3023371456986574,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004987068300119173,
|
|
"loss": 6.9845,
|
|
"mean_token_accuracy": 0.11879340037703515,
|
|
"num_tokens": 2568840.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.427999258041382,
|
|
"epoch": 1.3073097961213327,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986646830356792,
|
|
"loss": 7.0719,
|
|
"mean_token_accuracy": 0.11699894666671753,
|
|
"num_tokens": 2578566.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.310933828353882,
|
|
"epoch": 1.312282446544008,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004986218622456423,
|
|
"loss": 7.0501,
|
|
"mean_token_accuracy": 0.11614587306976318,
|
|
"num_tokens": 2589853.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.3128269672393795,
|
|
"epoch": 1.3172550969666832,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000498578367770809,
|
|
"loss": 7.0068,
|
|
"mean_token_accuracy": 0.12249249666929245,
|
|
"num_tokens": 2600848.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.256073331832885,
|
|
"epoch": 1.3222277473893587,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004985341997422111,
|
|
"loss": 6.963,
|
|
"mean_token_accuracy": 0.12099890038371086,
|
|
"num_tokens": 2610730.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.297663879394531,
|
|
"epoch": 1.327200397812034,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004984893582929093,
|
|
"loss": 7.0452,
|
|
"mean_token_accuracy": 0.11725059002637864,
|
|
"num_tokens": 2621129.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.355408573150635,
|
|
"epoch": 1.3321730482347092,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004984438435579933,
|
|
"loss": 6.982,
|
|
"mean_token_accuracy": 0.12046277597546577,
|
|
"num_tokens": 2630306.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.296045589447021,
|
|
"epoch": 1.3371456986573844,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983976556745812,
|
|
"loss": 7.0823,
|
|
"mean_token_accuracy": 0.11265691295266152,
|
|
"num_tokens": 2639874.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.283325958251953,
|
|
"epoch": 1.3421183490800597,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000498350794781819,
|
|
"loss": 7.0678,
|
|
"mean_token_accuracy": 0.11373703181743622,
|
|
"num_tokens": 2650340.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.207629680633545,
|
|
"epoch": 1.347090999502735,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004983032610208801,
|
|
"loss": 6.8353,
|
|
"mean_token_accuracy": 0.12716119810938836,
|
|
"num_tokens": 2659541.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.286678600311279,
|
|
"epoch": 1.3520636499254102,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004982550545349649,
|
|
"loss": 7.0215,
|
|
"mean_token_accuracy": 0.1187352031469345,
|
|
"num_tokens": 2669151.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.200907421112061,
|
|
"epoch": 1.3570363003480854,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004982061754693008,
|
|
"loss": 6.9129,
|
|
"mean_token_accuracy": 0.12817951813340187,
|
|
"num_tokens": 2678867.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.214226150512696,
|
|
"epoch": 1.362008950770761,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004981566239711412,
|
|
"loss": 7.015,
|
|
"mean_token_accuracy": 0.11664054766297341,
|
|
"num_tokens": 2687572.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.275300598144531,
|
|
"epoch": 1.3669816011934361,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004981064001897651,
|
|
"loss": 6.9533,
|
|
"mean_token_accuracy": 0.12162196189165116,
|
|
"num_tokens": 2696922.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.232679748535157,
|
|
"epoch": 1.3719542516161114,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004980555042764772,
|
|
"loss": 6.9895,
|
|
"mean_token_accuracy": 0.1217948891222477,
|
|
"num_tokens": 2707465.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.308532428741455,
|
|
"epoch": 1.3769269020387866,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004980039363846069,
|
|
"loss": 6.913,
|
|
"mean_token_accuracy": 0.12258979007601738,
|
|
"num_tokens": 2718708.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.277252340316773,
|
|
"epoch": 1.381899552461462,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000497951696669508,
|
|
"loss": 6.9481,
|
|
"mean_token_accuracy": 0.12027385532855987,
|
|
"num_tokens": 2728862.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.274434232711792,
|
|
"epoch": 1.3868722028841374,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004978987852885579,
|
|
"loss": 6.9239,
|
|
"mean_token_accuracy": 0.1204326257109642,
|
|
"num_tokens": 2738546.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.141081762313843,
|
|
"epoch": 1.3918448533068126,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004978452024011582,
|
|
"loss": 6.8723,
|
|
"mean_token_accuracy": 0.12342199310660362,
|
|
"num_tokens": 2747885.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.203327798843384,
|
|
"epoch": 1.3968175037294879,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000497790948168733,
|
|
"loss": 6.9071,
|
|
"mean_token_accuracy": 0.12794516757130622,
|
|
"num_tokens": 2756851.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.273374366760254,
|
|
"epoch": 1.4017901541521631,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004977360227547288,
|
|
"loss": 6.9432,
|
|
"mean_token_accuracy": 0.12073535323143006,
|
|
"num_tokens": 2766309.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.190575027465821,
|
|
"epoch": 1.4067628045748384,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004976804263246146,
|
|
"loss": 6.933,
|
|
"mean_token_accuracy": 0.12165748327970505,
|
|
"num_tokens": 2775679.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.199769639968872,
|
|
"epoch": 1.4117354549975136,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004976241590458803,
|
|
"loss": 6.9014,
|
|
"mean_token_accuracy": 0.11840908005833625,
|
|
"num_tokens": 2784610.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.220741128921508,
|
|
"epoch": 1.4167081054201889,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004975672210880372,
|
|
"loss": 6.9827,
|
|
"mean_token_accuracy": 0.12454970851540566,
|
|
"num_tokens": 2793969.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.191783380508423,
|
|
"epoch": 1.4216807558428641,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004975096126226171,
|
|
"loss": 6.8273,
|
|
"mean_token_accuracy": 0.12485689148306847,
|
|
"num_tokens": 2803096.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.134825086593628,
|
|
"epoch": 1.4266534062655396,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004974513338231715,
|
|
"loss": 6.8946,
|
|
"mean_token_accuracy": 0.12695471122860907,
|
|
"num_tokens": 2812320.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.318533086776734,
|
|
"epoch": 1.4316260566882149,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004973923848652715,
|
|
"loss": 6.9699,
|
|
"mean_token_accuracy": 0.1234718531370163,
|
|
"num_tokens": 2821393.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.124846076965332,
|
|
"epoch": 1.43659870711089,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004973327659265073,
|
|
"loss": 6.9346,
|
|
"mean_token_accuracy": 0.12769981995224952,
|
|
"num_tokens": 2831182.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.24030909538269,
|
|
"epoch": 1.4415713575335654,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004972724771864874,
|
|
"loss": 6.9004,
|
|
"mean_token_accuracy": 0.12435437738895416,
|
|
"num_tokens": 2840429.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.124129819869995,
|
|
"epoch": 1.4465440079562406,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004972115188268378,
|
|
"loss": 6.8795,
|
|
"mean_token_accuracy": 0.13040579184889795,
|
|
"num_tokens": 2849443.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.2121350288391115,
|
|
"epoch": 1.451516658378916,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004971498910312024,
|
|
"loss": 6.8697,
|
|
"mean_token_accuracy": 0.1304308533668518,
|
|
"num_tokens": 2858908.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.27411732673645,
|
|
"epoch": 1.4564893088015913,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004970875939852412,
|
|
"loss": 6.9543,
|
|
"mean_token_accuracy": 0.1235989436507225,
|
|
"num_tokens": 2869615.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.167358160018921,
|
|
"epoch": 1.4614619592242666,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497024627876631,
|
|
"loss": 6.8862,
|
|
"mean_token_accuracy": 0.12236595377326012,
|
|
"num_tokens": 2879007.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.165453147888184,
|
|
"epoch": 1.4664346096469418,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004969609928950639,
|
|
"loss": 6.7867,
|
|
"mean_token_accuracy": 0.12878735810518266,
|
|
"num_tokens": 2887527.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.164786005020142,
|
|
"epoch": 1.471407260069617,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004968966892322468,
|
|
"loss": 6.8409,
|
|
"mean_token_accuracy": 0.1268341325223446,
|
|
"num_tokens": 2897137.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.142514228820801,
|
|
"epoch": 1.4763799104922923,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004968317170819018,
|
|
"loss": 6.9224,
|
|
"mean_token_accuracy": 0.12766365706920624,
|
|
"num_tokens": 2908346.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.196111488342285,
|
|
"epoch": 1.4813525609149676,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004967660766397642,
|
|
"loss": 6.9472,
|
|
"mean_token_accuracy": 0.1233054369688034,
|
|
"num_tokens": 2918438.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.271194982528686,
|
|
"epoch": 1.4863252113376428,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004966997681035829,
|
|
"loss": 7.0489,
|
|
"mean_token_accuracy": 0.11800875365734101,
|
|
"num_tokens": 2928096.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.1524786949157715,
|
|
"epoch": 1.4912978617603183,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004966327916731195,
|
|
"loss": 6.8858,
|
|
"mean_token_accuracy": 0.12676455900073053,
|
|
"num_tokens": 2938150.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.122628211975098,
|
|
"epoch": 1.4962705121829936,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004965651475501476,
|
|
"loss": 6.8568,
|
|
"mean_token_accuracy": 0.12637797221541405,
|
|
"num_tokens": 2949410.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.167972421646118,
|
|
"epoch": 1.5012431626056688,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004964968359384524,
|
|
"loss": 6.8297,
|
|
"mean_token_accuracy": 0.13483862057328225,
|
|
"num_tokens": 2958564.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.077683067321777,
|
|
"epoch": 1.506215813028344,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004964278570438299,
|
|
"loss": 6.8817,
|
|
"mean_token_accuracy": 0.1293439321219921,
|
|
"num_tokens": 2968750.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.195564079284668,
|
|
"epoch": 1.5111884634510195,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004963582110740865,
|
|
"loss": 6.8386,
|
|
"mean_token_accuracy": 0.13061806932091713,
|
|
"num_tokens": 2978397.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.144682598114014,
|
|
"epoch": 1.5161611138736948,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004962878982390381,
|
|
"loss": 6.965,
|
|
"mean_token_accuracy": 0.11757178157567978,
|
|
"num_tokens": 2987625.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.171010541915893,
|
|
"epoch": 1.52113376429637,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004962169187505097,
|
|
"loss": 6.9069,
|
|
"mean_token_accuracy": 0.12990204840898514,
|
|
"num_tokens": 2997266.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.161147403717041,
|
|
"epoch": 1.5261064147190453,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004961452728223343,
|
|
"loss": 6.9191,
|
|
"mean_token_accuracy": 0.12262054905295372,
|
|
"num_tokens": 3007474.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.283528518676758,
|
|
"epoch": 1.5310790651417205,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004960729606703535,
|
|
"loss": 6.8682,
|
|
"mean_token_accuracy": 0.12481922805309295,
|
|
"num_tokens": 3016270.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.0521135330200195,
|
|
"epoch": 1.5360517155643958,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004959999825124147,
|
|
"loss": 6.8737,
|
|
"mean_token_accuracy": 0.1247307188808918,
|
|
"num_tokens": 3027397.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.23290376663208,
|
|
"epoch": 1.541024365987071,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000495926338568373,
|
|
"loss": 6.8929,
|
|
"mean_token_accuracy": 0.12644692212343217,
|
|
"num_tokens": 3037324.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.104544448852539,
|
|
"epoch": 1.5459970164097463,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004958520290600884,
|
|
"loss": 6.8639,
|
|
"mean_token_accuracy": 0.12832380905747415,
|
|
"num_tokens": 3048315.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.089476490020752,
|
|
"epoch": 1.5509696668324215,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004957770542114262,
|
|
"loss": 6.7832,
|
|
"mean_token_accuracy": 0.131186331063509,
|
|
"num_tokens": 3057875.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.199433183670044,
|
|
"epoch": 1.555942317255097,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004957014142482563,
|
|
"loss": 6.8823,
|
|
"mean_token_accuracy": 0.12605154365301133,
|
|
"num_tokens": 3068006.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.11689567565918,
|
|
"epoch": 1.5609149676777723,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004956251093984521,
|
|
"loss": 6.8874,
|
|
"mean_token_accuracy": 0.13267903551459312,
|
|
"num_tokens": 3079450.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.067438125610352,
|
|
"epoch": 1.5658876181004475,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004955481398918902,
|
|
"loss": 6.6779,
|
|
"mean_token_accuracy": 0.14047748744487762,
|
|
"num_tokens": 3088385.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.163371229171753,
|
|
"epoch": 1.5708602685231228,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004954705059604495,
|
|
"loss": 6.8313,
|
|
"mean_token_accuracy": 0.12182854935526848,
|
|
"num_tokens": 3097854.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.2161249160766605,
|
|
"epoch": 1.5758329189457982,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004953922078380104,
|
|
"loss": 6.9489,
|
|
"mean_token_accuracy": 0.12091099992394447,
|
|
"num_tokens": 3107877.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.058309602737427,
|
|
"epoch": 1.5808055693684735,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004953132457604544,
|
|
"loss": 6.8951,
|
|
"mean_token_accuracy": 0.1263230949640274,
|
|
"num_tokens": 3117967.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.053575754165649,
|
|
"epoch": 1.5857782197911487,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004952336199656632,
|
|
"loss": 6.7892,
|
|
"mean_token_accuracy": 0.12636456564068793,
|
|
"num_tokens": 3129827.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.151377630233765,
|
|
"epoch": 1.590750870213824,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004951533306935181,
|
|
"loss": 6.7993,
|
|
"mean_token_accuracy": 0.128205543756485,
|
|
"num_tokens": 3138779.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.978274393081665,
|
|
"epoch": 1.5957235206364992,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004950723781858991,
|
|
"loss": 6.8264,
|
|
"mean_token_accuracy": 0.13113004490733146,
|
|
"num_tokens": 3148741.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.116631937026978,
|
|
"epoch": 1.6006961710591745,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004949907626866844,
|
|
"loss": 6.8637,
|
|
"mean_token_accuracy": 0.13250831812620162,
|
|
"num_tokens": 3158964.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.00434422492981,
|
|
"epoch": 1.6056688214818498,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004949084844417493,
|
|
"loss": 6.6934,
|
|
"mean_token_accuracy": 0.13387084379792213,
|
|
"num_tokens": 3168347.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.077120351791382,
|
|
"epoch": 1.610641471904525,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004948255436989657,
|
|
"loss": 6.8875,
|
|
"mean_token_accuracy": 0.12550292983651162,
|
|
"num_tokens": 3178276.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.056673669815064,
|
|
"epoch": 1.6156141223272003,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004947419407082017,
|
|
"loss": 6.8322,
|
|
"mean_token_accuracy": 0.12148996368050576,
|
|
"num_tokens": 3186975.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.0625499248504635,
|
|
"epoch": 1.6205867727498757,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004946576757213202,
|
|
"loss": 6.6879,
|
|
"mean_token_accuracy": 0.14427319616079332,
|
|
"num_tokens": 3197311.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.04554591178894,
|
|
"epoch": 1.625559423172551,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004945727489921784,
|
|
"loss": 6.8213,
|
|
"mean_token_accuracy": 0.12860012352466582,
|
|
"num_tokens": 3208534.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.157459926605225,
|
|
"epoch": 1.6305320735952262,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004944871607766273,
|
|
"loss": 6.7724,
|
|
"mean_token_accuracy": 0.13342422023415565,
|
|
"num_tokens": 3217440.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.10877947807312,
|
|
"epoch": 1.6355047240179017,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004944009113325104,
|
|
"loss": 6.9024,
|
|
"mean_token_accuracy": 0.1263715833425522,
|
|
"num_tokens": 3225858.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.9752678871154785,
|
|
"epoch": 1.640477374440577,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004943140009196635,
|
|
"loss": 6.6761,
|
|
"mean_token_accuracy": 0.14351459145545958,
|
|
"num_tokens": 3235921.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.0338750839233395,
|
|
"epoch": 1.6454500248632522,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004942264297999135,
|
|
"loss": 6.7436,
|
|
"mean_token_accuracy": 0.13987542390823365,
|
|
"num_tokens": 3245123.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.107439851760864,
|
|
"epoch": 1.6504226752859275,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004941381982370778,
|
|
"loss": 6.7838,
|
|
"mean_token_accuracy": 0.13506464436650276,
|
|
"num_tokens": 3254246.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.082689619064331,
|
|
"epoch": 1.6553953257086027,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004940493064969633,
|
|
"loss": 6.7581,
|
|
"mean_token_accuracy": 0.12909987792372704,
|
|
"num_tokens": 3263602.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.0795430660247805,
|
|
"epoch": 1.660367976131278,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000493959754847366,
|
|
"loss": 6.9324,
|
|
"mean_token_accuracy": 0.12523805052042009,
|
|
"num_tokens": 3274430.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.031734418869019,
|
|
"epoch": 1.6653406265539532,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004938695435580698,
|
|
"loss": 6.6661,
|
|
"mean_token_accuracy": 0.13582849130034447,
|
|
"num_tokens": 3283587.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.0310241222381595,
|
|
"epoch": 1.6703132769766285,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004937786729008459,
|
|
"loss": 6.8373,
|
|
"mean_token_accuracy": 0.11871433779597282,
|
|
"num_tokens": 3294165.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.075614786148071,
|
|
"epoch": 1.6752859273993037,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004936871431494517,
|
|
"loss": 6.8026,
|
|
"mean_token_accuracy": 0.132364172488451,
|
|
"num_tokens": 3302756.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.035198593139649,
|
|
"epoch": 1.680258577821979,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004935949545796307,
|
|
"loss": 6.7383,
|
|
"mean_token_accuracy": 0.13242980763316153,
|
|
"num_tokens": 3313108.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.0880231857299805,
|
|
"epoch": 1.6852312282446544,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935021074691106,
|
|
"loss": 6.8111,
|
|
"mean_token_accuracy": 0.1307057686150074,
|
|
"num_tokens": 3323008.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.031533050537109,
|
|
"epoch": 1.6902038786673297,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004934086020976032,
|
|
"loss": 6.737,
|
|
"mean_token_accuracy": 0.13293734863400458,
|
|
"num_tokens": 3332490.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.939979791641235,
|
|
"epoch": 1.695176529090005,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004933144387468037,
|
|
"loss": 6.777,
|
|
"mean_token_accuracy": 0.1284017875790596,
|
|
"num_tokens": 3343069.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.02036862373352,
|
|
"epoch": 1.7001491795126804,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004932196177003891,
|
|
"loss": 6.8705,
|
|
"mean_token_accuracy": 0.13063741773366927,
|
|
"num_tokens": 3351834.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.201393175125122,
|
|
"epoch": 1.7051218299353557,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000493124139244018,
|
|
"loss": 6.7634,
|
|
"mean_token_accuracy": 0.1267462633550167,
|
|
"num_tokens": 3360915.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.997991943359375,
|
|
"epoch": 1.710094480358031,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004930280036653294,
|
|
"loss": 6.7547,
|
|
"mean_token_accuracy": 0.1312308818101883,
|
|
"num_tokens": 3370184.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.991842031478882,
|
|
"epoch": 1.7150671307807062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492931211253942,
|
|
"loss": 6.8364,
|
|
"mean_token_accuracy": 0.12366291210055351,
|
|
"num_tokens": 3379545.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.048074626922608,
|
|
"epoch": 1.7200397812033814,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004928337623014535,
|
|
"loss": 6.7862,
|
|
"mean_token_accuracy": 0.13503738194704057,
|
|
"num_tokens": 3390238.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.057003545761108,
|
|
"epoch": 1.7250124316260567,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000492735657101439,
|
|
"loss": 6.7912,
|
|
"mean_token_accuracy": 0.13303097039461137,
|
|
"num_tokens": 3400437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.988606405258179,
|
|
"epoch": 1.729985082048732,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004926368959494513,
|
|
"loss": 6.768,
|
|
"mean_token_accuracy": 0.1322055734694004,
|
|
"num_tokens": 3410982.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.101892518997192,
|
|
"epoch": 1.7349577324714072,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004925374791430186,
|
|
"loss": 6.873,
|
|
"mean_token_accuracy": 0.1290893539786339,
|
|
"num_tokens": 3420898.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.0329371929168705,
|
|
"epoch": 1.7399303828940824,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004924374069816448,
|
|
"loss": 6.7248,
|
|
"mean_token_accuracy": 0.13272165432572364,
|
|
"num_tokens": 3429144.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.968243074417114,
|
|
"epoch": 1.7449030333167577,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004923366797668081,
|
|
"loss": 6.7567,
|
|
"mean_token_accuracy": 0.13730168342590332,
|
|
"num_tokens": 3439427.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.095388031005859,
|
|
"epoch": 1.7498756837394331,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004922352978019597,
|
|
"loss": 6.7449,
|
|
"mean_token_accuracy": 0.13652047142386436,
|
|
"num_tokens": 3448218.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.058883094787598,
|
|
"epoch": 1.7548483341621084,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004921332613925239,
|
|
"loss": 6.8098,
|
|
"mean_token_accuracy": 0.13411547467112542,
|
|
"num_tokens": 3457656.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.090395736694336,
|
|
"epoch": 1.7598209845847836,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004920305708458964,
|
|
"loss": 6.777,
|
|
"mean_token_accuracy": 0.13345538526773454,
|
|
"num_tokens": 3467414.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.967651224136352,
|
|
"epoch": 1.7647936350074591,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004919272264714431,
|
|
"loss": 6.7212,
|
|
"mean_token_accuracy": 0.13109994381666185,
|
|
"num_tokens": 3477195.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.005810976028442,
|
|
"epoch": 1.7697662854301344,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004918232285805004,
|
|
"loss": 6.7991,
|
|
"mean_token_accuracy": 0.12487776428461075,
|
|
"num_tokens": 3487497.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.072210121154785,
|
|
"epoch": 1.7747389358528096,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004917185774863727,
|
|
"loss": 6.7683,
|
|
"mean_token_accuracy": 0.12752508446574212,
|
|
"num_tokens": 3497363.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.987413120269776,
|
|
"epoch": 1.7797115862754849,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004916132735043329,
|
|
"loss": 6.7762,
|
|
"mean_token_accuracy": 0.13657397255301476,
|
|
"num_tokens": 3506969.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.0449316024780275,
|
|
"epoch": 1.7846842366981601,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004915073169516204,
|
|
"loss": 6.7243,
|
|
"mean_token_accuracy": 0.13530478179454802,
|
|
"num_tokens": 3516418.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.944989109039307,
|
|
"epoch": 1.7896568871208354,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004914007081474406,
|
|
"loss": 6.6562,
|
|
"mean_token_accuracy": 0.13642907813191413,
|
|
"num_tokens": 3528023.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.036559677124023,
|
|
"epoch": 1.7946295375435106,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004912934474129641,
|
|
"loss": 6.7426,
|
|
"mean_token_accuracy": 0.12724815607070922,
|
|
"num_tokens": 3537946.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.9891579151153564,
|
|
"epoch": 1.7996021879661859,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004911855350713251,
|
|
"loss": 6.7398,
|
|
"mean_token_accuracy": 0.13197653666138648,
|
|
"num_tokens": 3547299.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.039433288574219,
|
|
"epoch": 1.8045748383888611,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004910769714476214,
|
|
"loss": 6.803,
|
|
"mean_token_accuracy": 0.1329810783267021,
|
|
"num_tokens": 3556075.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.974453401565552,
|
|
"epoch": 1.8095474888115366,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004909677568689122,
|
|
"loss": 6.7743,
|
|
"mean_token_accuracy": 0.1359911873936653,
|
|
"num_tokens": 3566314.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.997809982299804,
|
|
"epoch": 1.8145201392342118,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004908578916642183,
|
|
"loss": 6.6671,
|
|
"mean_token_accuracy": 0.1372392177581787,
|
|
"num_tokens": 3575095.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.998040199279785,
|
|
"epoch": 1.819492789656887,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004907473761645205,
|
|
"loss": 6.7679,
|
|
"mean_token_accuracy": 0.12976396828889847,
|
|
"num_tokens": 3584375.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.929082679748535,
|
|
"epoch": 1.8244654400795626,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004906362107027583,
|
|
"loss": 6.6429,
|
|
"mean_token_accuracy": 0.13891078382730485,
|
|
"num_tokens": 3594504.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.957910633087158,
|
|
"epoch": 1.8294380905022378,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004905243956138298,
|
|
"loss": 6.7883,
|
|
"mean_token_accuracy": 0.12719112262129784,
|
|
"num_tokens": 3604050.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.0381217956542965,
|
|
"epoch": 1.834410740924913,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004904119312345896,
|
|
"loss": 6.6852,
|
|
"mean_token_accuracy": 0.138168054074049,
|
|
"num_tokens": 3613388.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.980727005004883,
|
|
"epoch": 1.8393833913475883,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004902988179038492,
|
|
"loss": 6.864,
|
|
"mean_token_accuracy": 0.1205054223537445,
|
|
"num_tokens": 3623758.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.040994453430176,
|
|
"epoch": 1.8443560417702636,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000490185055962374,
|
|
"loss": 6.725,
|
|
"mean_token_accuracy": 0.13613900542259216,
|
|
"num_tokens": 3633040.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.923250293731689,
|
|
"epoch": 1.8493286921929388,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.0004900706457528845,
|
|
"loss": 6.6361,
|
|
"mean_token_accuracy": 0.13385234624147416,
|
|
"num_tokens": 3642980.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.012883281707763,
|
|
"epoch": 1.854301342615614,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004899555876200535,
|
|
"loss": 6.7982,
|
|
"mean_token_accuracy": 0.13437614887952803,
|
|
"num_tokens": 3653817.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.995367908477784,
|
|
"epoch": 1.8592739930382893,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004898398819105059,
|
|
"loss": 6.7398,
|
|
"mean_token_accuracy": 0.12992661371827124,
|
|
"num_tokens": 3664460.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.042414999008178,
|
|
"epoch": 1.8642466434609646,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004897235289728178,
|
|
"loss": 6.7557,
|
|
"mean_token_accuracy": 0.12437410354614258,
|
|
"num_tokens": 3674221.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.882666158676147,
|
|
"epoch": 1.8692192938836398,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004896065291575145,
|
|
"loss": 6.6562,
|
|
"mean_token_accuracy": 0.13729194477200507,
|
|
"num_tokens": 3683130.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.057892894744873,
|
|
"epoch": 1.8741919443063153,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004894888828170708,
|
|
"loss": 6.6045,
|
|
"mean_token_accuracy": 0.1372959293425083,
|
|
"num_tokens": 3694544.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.888378858566284,
|
|
"epoch": 1.8791645947289906,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004893705903059087,
|
|
"loss": 6.5839,
|
|
"mean_token_accuracy": 0.1371932953596115,
|
|
"num_tokens": 3704857.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.931747770309448,
|
|
"epoch": 1.8841372451516658,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004892516519803971,
|
|
"loss": 6.6791,
|
|
"mean_token_accuracy": 0.13744482696056365,
|
|
"num_tokens": 3714623.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.939040327072144,
|
|
"epoch": 1.8891098955743413,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004891320681988504,
|
|
"loss": 6.6753,
|
|
"mean_token_accuracy": 0.13932455331087112,
|
|
"num_tokens": 3723349.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.948206424713135,
|
|
"epoch": 1.8940825459970165,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004890118393215277,
|
|
"loss": 6.6438,
|
|
"mean_token_accuracy": 0.13924443125724792,
|
|
"num_tokens": 3732990.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.890056848526001,
|
|
"epoch": 1.8990551964196918,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004888909657106311,
|
|
"loss": 6.6316,
|
|
"mean_token_accuracy": 0.13604750409722327,
|
|
"num_tokens": 3742343.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.889602947235107,
|
|
"epoch": 1.904027846842367,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004887694477303055,
|
|
"loss": 6.6534,
|
|
"mean_token_accuracy": 0.1300087846815586,
|
|
"num_tokens": 3752128.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.939905405044556,
|
|
"epoch": 1.9090004972650423,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004886472857466368,
|
|
"loss": 6.5785,
|
|
"mean_token_accuracy": 0.13404396995902063,
|
|
"num_tokens": 3760431.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.90602216720581,
|
|
"epoch": 1.9139731476877175,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004885244801276509,
|
|
"loss": 6.6925,
|
|
"mean_token_accuracy": 0.13369300588965416,
|
|
"num_tokens": 3769623.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.837252616882324,
|
|
"epoch": 1.9189457981103928,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004884010312433131,
|
|
"loss": 6.6385,
|
|
"mean_token_accuracy": 0.1345611870288849,
|
|
"num_tokens": 3779383.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.014624786376953,
|
|
"epoch": 1.923918448533068,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048827693946552615,
|
|
"loss": 6.7744,
|
|
"mean_token_accuracy": 0.1251063071191311,
|
|
"num_tokens": 3788847.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.846893644332885,
|
|
"epoch": 1.9288910989557433,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048815220516813004,
|
|
"loss": 6.6404,
|
|
"mean_token_accuracy": 0.1399201013147831,
|
|
"num_tokens": 3798711.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.928702926635742,
|
|
"epoch": 1.9338637493784185,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004880268287269,
|
|
"loss": 6.6102,
|
|
"mean_token_accuracy": 0.1406691074371338,
|
|
"num_tokens": 3808837.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.9386796951293945,
|
|
"epoch": 1.938836399801094,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048790081051954605,
|
|
"loss": 6.5861,
|
|
"mean_token_accuracy": 0.13627308011054992,
|
|
"num_tokens": 3818261.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.9637360095977785,
|
|
"epoch": 1.9438090502237693,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004877741509257116,
|
|
"loss": 6.6477,
|
|
"mean_token_accuracy": 0.1298605188727379,
|
|
"num_tokens": 3828146.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.991808176040649,
|
|
"epoch": 1.9487817006464445,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004876468503269721,
|
|
"loss": 6.8629,
|
|
"mean_token_accuracy": 0.12371929138898849,
|
|
"num_tokens": 3838658.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.021228361129761,
|
|
"epoch": 1.95375435106912,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048751890910683434,
|
|
"loss": 6.6322,
|
|
"mean_token_accuracy": 0.13443570584058762,
|
|
"num_tokens": 3849097.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.900594758987427,
|
|
"epoch": 1.9587270014917952,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048739032765073487,
|
|
"loss": 6.7618,
|
|
"mean_token_accuracy": 0.1243375763297081,
|
|
"num_tokens": 3858645.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.9472784996032715,
|
|
"epoch": 1.9636996519144705,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048726110634603913,
|
|
"loss": 6.6418,
|
|
"mean_token_accuracy": 0.13881978690624236,
|
|
"num_tokens": 3869232.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.023511171340942,
|
|
"epoch": 1.9686723023371457,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048713124558204013,
|
|
"loss": 6.7395,
|
|
"mean_token_accuracy": 0.12681566774845124,
|
|
"num_tokens": 3879453.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.937405347824097,
|
|
"epoch": 1.973644952759821,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004870007457499573,
|
|
"loss": 6.6606,
|
|
"mean_token_accuracy": 0.13165320381522178,
|
|
"num_tokens": 3888257.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.962239980697632,
|
|
"epoch": 1.9786176031824962,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048686960724293537,
|
|
"loss": 6.6738,
|
|
"mean_token_accuracy": 0.14019001498818398,
|
|
"num_tokens": 3897980.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.833528661727906,
|
|
"epoch": 1.9835902536051715,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004867378304560432,
|
|
"loss": 6.5013,
|
|
"mean_token_accuracy": 0.14112157002091408,
|
|
"num_tokens": 3907479.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.905864000320435,
|
|
"epoch": 1.9885629040278467,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004866054157862723,
|
|
"loss": 6.7189,
|
|
"mean_token_accuracy": 0.13371711671352388,
|
|
"num_tokens": 3917249.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9885629040278467,
|
|
"eval_entropy": 6.75590797503423,
|
|
"eval_loss": 6.833089828491211,
|
|
"eval_mean_token_accuracy": 0.12698067444878788,
|
|
"eval_num_tokens": 3917249.0,
|
|
"eval_runtime": 1.9323,
|
|
"eval_samples_per_second": 1793.164,
|
|
"eval_steps_per_second": 224.598,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.8639627456665036,
|
|
"epoch": 1.993535554450522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004864723636325361,
|
|
"loss": 6.6244,
|
|
"mean_token_accuracy": 0.13904275372624397,
|
|
"num_tokens": 3928388.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.913655757904053,
|
|
"epoch": 1.9985082048731975,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048633867439566855,
|
|
"loss": 6.6671,
|
|
"mean_token_accuracy": 0.13446932137012482,
|
|
"num_tokens": 3937323.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.91440306769477,
|
|
"epoch": 2.002983590253605,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004862043484784227,
|
|
"loss": 6.5534,
|
|
"mean_token_accuracy": 0.13889687259991965,
|
|
"num_tokens": 3946815.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.9424295902252195,
|
|
"epoch": 2.0079562406762803,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048606938628546975,
|
|
"loss": 6.3361,
|
|
"mean_token_accuracy": 0.14214283972978592,
|
|
"num_tokens": 3956125.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.927752351760864,
|
|
"epoch": 2.0129288910989556,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004859337882233978,
|
|
"loss": 6.3751,
|
|
"mean_token_accuracy": 0.14679156690835954,
|
|
"num_tokens": 3966543.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.825806903839111,
|
|
"epoch": 2.017901541521631,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004857975547007106,
|
|
"loss": 6.3952,
|
|
"mean_token_accuracy": 0.13907834514975548,
|
|
"num_tokens": 3978200.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.863557815551758,
|
|
"epoch": 2.0228741919443065,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048566068612782595,
|
|
"loss": 6.2524,
|
|
"mean_token_accuracy": 0.15035446137189865,
|
|
"num_tokens": 3988331.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.9250030517578125,
|
|
"epoch": 2.0278468423669818,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048552318291707533,
|
|
"loss": 6.3107,
|
|
"mean_token_accuracy": 0.14710997119545938,
|
|
"num_tokens": 3998976.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.73673791885376,
|
|
"epoch": 2.032819492789657,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048538504548270177,
|
|
"loss": 6.3343,
|
|
"mean_token_accuracy": 0.14275099635124205,
|
|
"num_tokens": 4008354.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.902276277542114,
|
|
"epoch": 2.0377921432123323,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048524627424085913,
|
|
"loss": 6.2653,
|
|
"mean_token_accuracy": 0.1537691354751587,
|
|
"num_tokens": 4018045.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.7338885307312015,
|
|
"epoch": 2.0427647936350075,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048510686960961057,
|
|
"loss": 6.288,
|
|
"mean_token_accuracy": 0.1580418199300766,
|
|
"num_tokens": 4027884.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.787254667282104,
|
|
"epoch": 2.0477374440576828,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004849668320089276,
|
|
"loss": 6.3192,
|
|
"mean_token_accuracy": 0.14816607907414436,
|
|
"num_tokens": 4037866.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.899358558654785,
|
|
"epoch": 2.052710094480358,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048482616186068826,
|
|
"loss": 6.3327,
|
|
"mean_token_accuracy": 0.1437992498278618,
|
|
"num_tokens": 4048126.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.825502729415893,
|
|
"epoch": 2.0576827449030333,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048468485958867664,
|
|
"loss": 6.3641,
|
|
"mean_token_accuracy": 0.14440456554293632,
|
|
"num_tokens": 4058179.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.779832983016968,
|
|
"epoch": 2.0626553953257085,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048454292561858083,
|
|
"loss": 6.2366,
|
|
"mean_token_accuracy": 0.14441021382808686,
|
|
"num_tokens": 4067479.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.819252872467041,
|
|
"epoch": 2.0676280457483838,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004844003603779921,
|
|
"loss": 6.3375,
|
|
"mean_token_accuracy": 0.14714281931519507,
|
|
"num_tokens": 4076624.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.750158309936523,
|
|
"epoch": 2.072600696171059,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004842571642964036,
|
|
"loss": 6.2768,
|
|
"mean_token_accuracy": 0.15155254155397416,
|
|
"num_tokens": 4085752.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.786676120758057,
|
|
"epoch": 2.0775733465937343,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048411333780520876,
|
|
"loss": 6.3173,
|
|
"mean_token_accuracy": 0.15306983888149261,
|
|
"num_tokens": 4095026.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.806438446044922,
|
|
"epoch": 2.0825459970164095,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048396888133770034,
|
|
"loss": 6.3211,
|
|
"mean_token_accuracy": 0.14878666996955872,
|
|
"num_tokens": 4104589.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.761303520202636,
|
|
"epoch": 2.087518647439085,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048382379532906897,
|
|
"loss": 6.274,
|
|
"mean_token_accuracy": 0.14757272005081176,
|
|
"num_tokens": 4114550.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.868024730682373,
|
|
"epoch": 2.0924912978617605,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004836780802164016,
|
|
"loss": 6.3667,
|
|
"mean_token_accuracy": 0.15075910836458206,
|
|
"num_tokens": 4125491.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.74778037071228,
|
|
"epoch": 2.0974639482844357,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048353173643868064,
|
|
"loss": 6.2956,
|
|
"mean_token_accuracy": 0.14790154695510865,
|
|
"num_tokens": 4134599.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.695480298995972,
|
|
"epoch": 2.102436598707111,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004833847644367826,
|
|
"loss": 6.2259,
|
|
"mean_token_accuracy": 0.15309807807207107,
|
|
"num_tokens": 4143640.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.817388820648193,
|
|
"epoch": 2.107409249129786,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048323716465347615,
|
|
"loss": 6.315,
|
|
"mean_token_accuracy": 0.1500886969268322,
|
|
"num_tokens": 4152558.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.687919664382934,
|
|
"epoch": 2.1123818995524615,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004830889375334215,
|
|
"loss": 6.1499,
|
|
"mean_token_accuracy": 0.16754205971956254,
|
|
"num_tokens": 4162238.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.783289813995362,
|
|
"epoch": 2.1173545499751367,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048294008352316875,
|
|
"loss": 6.2928,
|
|
"mean_token_accuracy": 0.14750559329986573,
|
|
"num_tokens": 4171437.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.699970245361328,
|
|
"epoch": 2.122327200397812,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004827906030711564,
|
|
"loss": 6.2738,
|
|
"mean_token_accuracy": 0.15120746046304703,
|
|
"num_tokens": 4181584.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.673524284362793,
|
|
"epoch": 2.1272998508204872,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048264049662771053,
|
|
"loss": 6.1864,
|
|
"mean_token_accuracy": 0.15439680516719817,
|
|
"num_tokens": 4191756.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.659884643554688,
|
|
"epoch": 2.1322725012431625,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048248976464504266,
|
|
"loss": 6.2118,
|
|
"mean_token_accuracy": 0.1581698775291443,
|
|
"num_tokens": 4202864.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.887399816513062,
|
|
"epoch": 2.1372451516658377,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048233840757724926,
|
|
"loss": 6.3022,
|
|
"mean_token_accuracy": 0.14388342201709747,
|
|
"num_tokens": 4212839.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.71376314163208,
|
|
"epoch": 2.142217802088513,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048218642588030943,
|
|
"loss": 6.318,
|
|
"mean_token_accuracy": 0.14816687628626823,
|
|
"num_tokens": 4223150.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.729928827285766,
|
|
"epoch": 2.1471904525111887,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048203382001208453,
|
|
"loss": 6.3223,
|
|
"mean_token_accuracy": 0.15376662462949753,
|
|
"num_tokens": 4232206.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.815326738357544,
|
|
"epoch": 2.152163102933864,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048188059043231585,
|
|
"loss": 6.2932,
|
|
"mean_token_accuracy": 0.15043203681707382,
|
|
"num_tokens": 4240992.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.716480731964111,
|
|
"epoch": 2.157135753356539,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048172673760262414,
|
|
"loss": 6.2908,
|
|
"mean_token_accuracy": 0.14839086160063744,
|
|
"num_tokens": 4250371.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.782298517227173,
|
|
"epoch": 2.1621084037792144,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048157226198650735,
|
|
"loss": 6.3298,
|
|
"mean_token_accuracy": 0.1501259073615074,
|
|
"num_tokens": 4260606.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.744562005996704,
|
|
"epoch": 2.1670810542018897,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004814171640493398,
|
|
"loss": 6.3054,
|
|
"mean_token_accuracy": 0.14946358352899553,
|
|
"num_tokens": 4269753.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.638996982574463,
|
|
"epoch": 2.172053704624565,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004812614442583708,
|
|
"loss": 6.2586,
|
|
"mean_token_accuracy": 0.1536099910736084,
|
|
"num_tokens": 4279485.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.887175607681274,
|
|
"epoch": 2.17702635504724,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004811051030827227,
|
|
"loss": 6.325,
|
|
"mean_token_accuracy": 0.14802768975496292,
|
|
"num_tokens": 4289491.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.701043319702149,
|
|
"epoch": 2.1819990054699154,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048094814099339006,
|
|
"loss": 6.3413,
|
|
"mean_token_accuracy": 0.15326175689697266,
|
|
"num_tokens": 4300536.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.8160816669464115,
|
|
"epoch": 2.1869716558925907,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004807905584632381,
|
|
"loss": 6.3969,
|
|
"mean_token_accuracy": 0.13592732772231103,
|
|
"num_tokens": 4311804.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.897786474227905,
|
|
"epoch": 2.191944306315266,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048063235596700087,
|
|
"loss": 6.3788,
|
|
"mean_token_accuracy": 0.14074521511793137,
|
|
"num_tokens": 4320662.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.748007535934448,
|
|
"epoch": 2.196916956737941,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048047353398128056,
|
|
"loss": 6.3844,
|
|
"mean_token_accuracy": 0.1427532948553562,
|
|
"num_tokens": 4330414.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.6975822925567625,
|
|
"epoch": 2.2018896071606164,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004803140929845452,
|
|
"loss": 6.332,
|
|
"mean_token_accuracy": 0.14934077560901643,
|
|
"num_tokens": 4340478.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.747043418884277,
|
|
"epoch": 2.2068622575832917,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048015403345712796,
|
|
"loss": 6.2831,
|
|
"mean_token_accuracy": 0.14892816692590713,
|
|
"num_tokens": 4350688.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.596918249130249,
|
|
"epoch": 2.2118349080059674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004799933558812252,
|
|
"loss": 6.1824,
|
|
"mean_token_accuracy": 0.16453079953789712,
|
|
"num_tokens": 4360554.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.714308595657348,
|
|
"epoch": 2.2168075584286426,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047983206074089546,
|
|
"loss": 6.2314,
|
|
"mean_token_accuracy": 0.15688856244087218,
|
|
"num_tokens": 4370723.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.7229101181030275,
|
|
"epoch": 2.221780208851318,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004796701485220576,
|
|
"loss": 6.3144,
|
|
"mean_token_accuracy": 0.14810621291399,
|
|
"num_tokens": 4380383.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.73874683380127,
|
|
"epoch": 2.226752859273993,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004795076197124893,
|
|
"loss": 6.432,
|
|
"mean_token_accuracy": 0.1337784394621849,
|
|
"num_tokens": 4390408.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.718470239639283,
|
|
"epoch": 2.2317255096966684,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047934447480182626,
|
|
"loss": 6.3019,
|
|
"mean_token_accuracy": 0.150127013027668,
|
|
"num_tokens": 4400043.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.760151386260986,
|
|
"epoch": 2.2366981601193436,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004791807142815598,
|
|
"loss": 6.3597,
|
|
"mean_token_accuracy": 0.14522059112787247,
|
|
"num_tokens": 4409420.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.660960578918457,
|
|
"epoch": 2.241670810542019,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047901633864503614,
|
|
"loss": 6.2334,
|
|
"mean_token_accuracy": 0.15857069939374924,
|
|
"num_tokens": 4418742.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.739702606201172,
|
|
"epoch": 2.246643460964694,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047885134838745433,
|
|
"loss": 6.2529,
|
|
"mean_token_accuracy": 0.15395656526088713,
|
|
"num_tokens": 4427921.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.678990364074707,
|
|
"epoch": 2.2516161113873694,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004786857440058654,
|
|
"loss": 6.3134,
|
|
"mean_token_accuracy": 0.14815936386585235,
|
|
"num_tokens": 4438029.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.640738534927368,
|
|
"epoch": 2.2565887618100446,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004785195259991701,
|
|
"loss": 6.266,
|
|
"mean_token_accuracy": 0.1550798997282982,
|
|
"num_tokens": 4446967.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.753130912780762,
|
|
"epoch": 2.26156141223272,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000478352694868118,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.1522296339273453,
|
|
"num_tokens": 4456469.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.671848201751709,
|
|
"epoch": 2.266534062655395,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004781852511153058,
|
|
"loss": 6.362,
|
|
"mean_token_accuracy": 0.15018730834126473,
|
|
"num_tokens": 4465949.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.708608818054199,
|
|
"epoch": 2.271506713078071,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047801719524517573,
|
|
"loss": 6.1672,
|
|
"mean_token_accuracy": 0.1552305981516838,
|
|
"num_tokens": 4475460.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.72004017829895,
|
|
"epoch": 2.276479363500746,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047784852776401403,
|
|
"loss": 6.3038,
|
|
"mean_token_accuracy": 0.150562085211277,
|
|
"num_tokens": 4485540.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.6739191055297855,
|
|
"epoch": 2.2814520139234213,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004776792491799495,
|
|
"loss": 6.2782,
|
|
"mean_token_accuracy": 0.15366951674222945,
|
|
"num_tokens": 4495545.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.6869548797607425,
|
|
"epoch": 2.2864246643460966,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047750936000295213,
|
|
"loss": 6.2302,
|
|
"mean_token_accuracy": 0.15376197397708893,
|
|
"num_tokens": 4505527.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.691086006164551,
|
|
"epoch": 2.291397314768772,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047733886074483115,
|
|
"loss": 6.2321,
|
|
"mean_token_accuracy": 0.15552613958716394,
|
|
"num_tokens": 4515366.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.692183542251587,
|
|
"epoch": 2.296369965191447,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004771677519192339,
|
|
"loss": 6.3229,
|
|
"mean_token_accuracy": 0.1504087947309017,
|
|
"num_tokens": 4525273.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.712653255462646,
|
|
"epoch": 2.3013426156141223,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004769960340416439,
|
|
"loss": 6.3019,
|
|
"mean_token_accuracy": 0.14406404718756677,
|
|
"num_tokens": 4533568.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.681150341033936,
|
|
"epoch": 2.3063152660367976,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047682370762937976,
|
|
"loss": 6.4425,
|
|
"mean_token_accuracy": 0.14237538427114488,
|
|
"num_tokens": 4544508.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.780780839920044,
|
|
"epoch": 2.311287916459473,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047665077320159315,
|
|
"loss": 6.2096,
|
|
"mean_token_accuracy": 0.14799811244010924,
|
|
"num_tokens": 4554586.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.660666656494141,
|
|
"epoch": 2.316260566882148,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047647723127926765,
|
|
"loss": 6.2311,
|
|
"mean_token_accuracy": 0.1516297921538353,
|
|
"num_tokens": 4563883.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.619417810440064,
|
|
"epoch": 2.3212332173048233,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004763030823852168,
|
|
"loss": 6.2773,
|
|
"mean_token_accuracy": 0.154463554918766,
|
|
"num_tokens": 4573675.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.735545587539673,
|
|
"epoch": 2.3262058677274986,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004761283270440829,
|
|
"loss": 6.2774,
|
|
"mean_token_accuracy": 0.14591650813817977,
|
|
"num_tokens": 4583617.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.712375450134277,
|
|
"epoch": 2.331178518150174,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004759529657823349,
|
|
"loss": 6.3073,
|
|
"mean_token_accuracy": 0.15352334082126617,
|
|
"num_tokens": 4593792.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.685167980194092,
|
|
"epoch": 2.336151168572849,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047577699912826755,
|
|
"loss": 6.3062,
|
|
"mean_token_accuracy": 0.14912531822919844,
|
|
"num_tokens": 4603324.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.685032606124878,
|
|
"epoch": 2.341123818995525,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047560042761199915,
|
|
"loss": 6.257,
|
|
"mean_token_accuracy": 0.15247461944818497,
|
|
"num_tokens": 4613509.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.6419229984283445,
|
|
"epoch": 2.3460964694182,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004754232517654704,
|
|
"loss": 6.2277,
|
|
"mean_token_accuracy": 0.15689927563071251,
|
|
"num_tokens": 4624282.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.538430404663086,
|
|
"epoch": 2.3510691198408753,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004752454721224424,
|
|
"loss": 6.1969,
|
|
"mean_token_accuracy": 0.15574504286050797,
|
|
"num_tokens": 4633217.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.636493587493897,
|
|
"epoch": 2.3560417702635505,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047506708921849535,
|
|
"loss": 6.2071,
|
|
"mean_token_accuracy": 0.1586204707622528,
|
|
"num_tokens": 4642824.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.651132869720459,
|
|
"epoch": 2.361014420686226,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004748881035910269,
|
|
"loss": 6.2709,
|
|
"mean_token_accuracy": 0.15237033516168594,
|
|
"num_tokens": 4651938.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.573367404937744,
|
|
"epoch": 2.365987071108901,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004747085157792504,
|
|
"loss": 6.2638,
|
|
"mean_token_accuracy": 0.14993397817015647,
|
|
"num_tokens": 4661695.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.731213855743408,
|
|
"epoch": 2.3709597215315763,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004745283263241933,
|
|
"loss": 6.3461,
|
|
"mean_token_accuracy": 0.14866317957639694,
|
|
"num_tokens": 4671354.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.6773193359375,
|
|
"epoch": 2.3759323719542516,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004743475357686957,
|
|
"loss": 6.1807,
|
|
"mean_token_accuracy": 0.16951350644230842,
|
|
"num_tokens": 4681795.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.669539308547973,
|
|
"epoch": 2.380905022376927,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047416614465740843,
|
|
"loss": 6.2838,
|
|
"mean_token_accuracy": 0.1591503992676735,
|
|
"num_tokens": 4691229.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.642325592041016,
|
|
"epoch": 2.385877672799602,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047398415353679167,
|
|
"loss": 6.371,
|
|
"mean_token_accuracy": 0.15041485130786897,
|
|
"num_tokens": 4701880.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.738368701934815,
|
|
"epoch": 2.3908503232222773,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004738015629551132,
|
|
"loss": 6.2715,
|
|
"mean_token_accuracy": 0.14913799613714218,
|
|
"num_tokens": 4712041.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.484014320373535,
|
|
"epoch": 2.395822973644953,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047361837346244655,
|
|
"loss": 6.0647,
|
|
"mean_token_accuracy": 0.16438253074884415,
|
|
"num_tokens": 4722576.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.676149940490722,
|
|
"epoch": 2.4007956240676283,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004734345856106698,
|
|
"loss": 6.3114,
|
|
"mean_token_accuracy": 0.1528013452887535,
|
|
"num_tokens": 4732569.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.725569725036621,
|
|
"epoch": 2.4057682744903035,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047325019995346355,
|
|
"loss": 6.3463,
|
|
"mean_token_accuracy": 0.14696133583784105,
|
|
"num_tokens": 4741973.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.7788084030151365,
|
|
"epoch": 2.4107409249129788,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004730652170463091,
|
|
"loss": 6.3094,
|
|
"mean_token_accuracy": 0.15152185410261154,
|
|
"num_tokens": 4751550.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.614918279647827,
|
|
"epoch": 2.415713575335654,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004728796374464875,
|
|
"loss": 6.2585,
|
|
"mean_token_accuracy": 0.15672131031751632,
|
|
"num_tokens": 4761620.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.639470005035401,
|
|
"epoch": 2.4206862257583293,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047269346171307716,
|
|
"loss": 6.2321,
|
|
"mean_token_accuracy": 0.1548467367887497,
|
|
"num_tokens": 4771893.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.671611499786377,
|
|
"epoch": 2.4256588761810045,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004725066904069523,
|
|
"loss": 6.2966,
|
|
"mean_token_accuracy": 0.1491101309657097,
|
|
"num_tokens": 4781094.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.569028520584107,
|
|
"epoch": 2.4306315266036798,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004723193240907814,
|
|
"loss": 6.2458,
|
|
"mean_token_accuracy": 0.15056999623775483,
|
|
"num_tokens": 4790337.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.640410470962524,
|
|
"epoch": 2.435604177026355,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004721313633290257,
|
|
"loss": 6.3033,
|
|
"mean_token_accuracy": 0.15093883872032166,
|
|
"num_tokens": 4801758.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.720381593704223,
|
|
"epoch": 2.4405768274490303,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000471942808687937,
|
|
"loss": 6.3651,
|
|
"mean_token_accuracy": 0.1434055522084236,
|
|
"num_tokens": 4811040.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.632902383804321,
|
|
"epoch": 2.4455494778717055,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047175366073555654,
|
|
"loss": 6.224,
|
|
"mean_token_accuracy": 0.14985619932413102,
|
|
"num_tokens": 4821361.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.625610399246216,
|
|
"epoch": 2.4505221282943808,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004715639200417125,
|
|
"loss": 6.2734,
|
|
"mean_token_accuracy": 0.14472699761390687,
|
|
"num_tokens": 4830919.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.575866603851319,
|
|
"epoch": 2.455494778717056,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004713735871780192,
|
|
"loss": 6.0893,
|
|
"mean_token_accuracy": 0.17171268910169601,
|
|
"num_tokens": 4840364.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.6481630325317385,
|
|
"epoch": 2.4604674291397313,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004711826627178747,
|
|
"loss": 6.2747,
|
|
"mean_token_accuracy": 0.15398500710725785,
|
|
"num_tokens": 4849044.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.594645261764526,
|
|
"epoch": 2.4654400795624065,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047099114723645944,
|
|
"loss": 6.2523,
|
|
"mean_token_accuracy": 0.1518503412604332,
|
|
"num_tokens": 4859183.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.654210901260376,
|
|
"epoch": 2.470412729985082,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047079904131073434,
|
|
"loss": 6.2485,
|
|
"mean_token_accuracy": 0.15296704247593879,
|
|
"num_tokens": 4869625.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.679134511947632,
|
|
"epoch": 2.4753853804077575,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004706063455194389,
|
|
"loss": 6.2354,
|
|
"mean_token_accuracy": 0.1589968591928482,
|
|
"num_tokens": 4880089.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.590752124786377,
|
|
"epoch": 2.4803580308304327,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047041306044308986,
|
|
"loss": 6.3292,
|
|
"mean_token_accuracy": 0.1499726079404354,
|
|
"num_tokens": 4890673.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.587073469161988,
|
|
"epoch": 2.485330681253108,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004702191866639792,
|
|
"loss": 6.179,
|
|
"mean_token_accuracy": 0.15413953810930253,
|
|
"num_tokens": 4899444.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.649888324737549,
|
|
"epoch": 2.490303331675783,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004700247247661726,
|
|
"loss": 6.3354,
|
|
"mean_token_accuracy": 0.1461266316473484,
|
|
"num_tokens": 4909186.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.532314920425415,
|
|
"epoch": 2.4952759820984585,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004698296753355072,
|
|
"loss": 6.1188,
|
|
"mean_token_accuracy": 0.16991740316152573,
|
|
"num_tokens": 4918886.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.650052070617676,
|
|
"epoch": 2.5002486325211337,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004696340389595903,
|
|
"loss": 6.3121,
|
|
"mean_token_accuracy": 0.15100516825914384,
|
|
"num_tokens": 4929157.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.6126251220703125,
|
|
"epoch": 2.505221282943809,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046943781622779733,
|
|
"loss": 6.1548,
|
|
"mean_token_accuracy": 0.1571856901049614,
|
|
"num_tokens": 4938127.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.6133850574493405,
|
|
"epoch": 2.510193933366484,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046924100773127055,
|
|
"loss": 6.1624,
|
|
"mean_token_accuracy": 0.1639612466096878,
|
|
"num_tokens": 4947849.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.546055746078491,
|
|
"epoch": 2.5151665837891595,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046904361406291633,
|
|
"loss": 6.224,
|
|
"mean_token_accuracy": 0.1560969039797783,
|
|
"num_tokens": 4957046.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.5485820293426515,
|
|
"epoch": 2.520139234211835,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004688456358174044,
|
|
"loss": 6.2604,
|
|
"mean_token_accuracy": 0.15628362596035003,
|
|
"num_tokens": 4967128.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.687307357788086,
|
|
"epoch": 2.5251118846345104,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004686470735911653,
|
|
"loss": 6.3171,
|
|
"mean_token_accuracy": 0.149687460064888,
|
|
"num_tokens": 4976817.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.674047803878784,
|
|
"epoch": 2.5300845350571857,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004684479279823891,
|
|
"loss": 6.3028,
|
|
"mean_token_accuracy": 0.1542089104652405,
|
|
"num_tokens": 4985908.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.630443477630616,
|
|
"epoch": 2.535057185479861,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004682481995910232,
|
|
"loss": 6.1957,
|
|
"mean_token_accuracy": 0.1554957613348961,
|
|
"num_tokens": 4995218.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.599108457565308,
|
|
"epoch": 2.540029835902536,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046804788901877084,
|
|
"loss": 6.3413,
|
|
"mean_token_accuracy": 0.14961577951908112,
|
|
"num_tokens": 5004197.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.6687071323394775,
|
|
"epoch": 2.5450024863252114,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046784699686908886,
|
|
"loss": 6.2752,
|
|
"mean_token_accuracy": 0.14543737843632698,
|
|
"num_tokens": 5015408.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.614528226852417,
|
|
"epoch": 2.5499751367478867,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046764552374718656,
|
|
"loss": 6.2581,
|
|
"mean_token_accuracy": 0.15023616328835487,
|
|
"num_tokens": 5025138.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.616063499450684,
|
|
"epoch": 2.554947787170562,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046744347026002315,
|
|
"loss": 6.2763,
|
|
"mean_token_accuracy": 0.1504281558096409,
|
|
"num_tokens": 5034638.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.642090654373169,
|
|
"epoch": 2.559920437593237,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004672408370163066,
|
|
"loss": 6.3783,
|
|
"mean_token_accuracy": 0.14809440523386003,
|
|
"num_tokens": 5044826.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.679381704330444,
|
|
"epoch": 2.5648930880159124,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000467037624626491,
|
|
"loss": 6.4388,
|
|
"mean_token_accuracy": 0.14451707750558854,
|
|
"num_tokens": 5054914.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.5161418437957765,
|
|
"epoch": 2.5698657384385877,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046683383370277554,
|
|
"loss": 6.1545,
|
|
"mean_token_accuracy": 0.16692716106772423,
|
|
"num_tokens": 5066689.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.631124019622803,
|
|
"epoch": 2.574838388861263,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046662946485910226,
|
|
"loss": 6.334,
|
|
"mean_token_accuracy": 0.1525747962296009,
|
|
"num_tokens": 5076170.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.6436363697052006,
|
|
"epoch": 2.579811039283938,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000466424518711154,
|
|
"loss": 6.195,
|
|
"mean_token_accuracy": 0.15618150681257248,
|
|
"num_tokens": 5085433.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.61197419166565,
|
|
"epoch": 2.5847836897066134,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004662189958763531,
|
|
"loss": 6.1874,
|
|
"mean_token_accuracy": 0.15478023886680603,
|
|
"num_tokens": 5095692.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.523416471481323,
|
|
"epoch": 2.5897563401292887,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004660128969738591,
|
|
"loss": 6.2744,
|
|
"mean_token_accuracy": 0.14609457030892373,
|
|
"num_tokens": 5105037.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.590671062469482,
|
|
"epoch": 2.594728990551964,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004658062226245668,
|
|
"loss": 6.2083,
|
|
"mean_token_accuracy": 0.15642113238573074,
|
|
"num_tokens": 5113868.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.744363069534302,
|
|
"epoch": 2.5997016409746396,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004655989734511052,
|
|
"loss": 6.2726,
|
|
"mean_token_accuracy": 0.15813782140612603,
|
|
"num_tokens": 5123136.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.569714879989624,
|
|
"epoch": 2.604674291397315,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004653911500778343,
|
|
"loss": 6.1441,
|
|
"mean_token_accuracy": 0.15831523686647414,
|
|
"num_tokens": 5133869.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.606399297714233,
|
|
"epoch": 2.60964694181999,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004651827531308443,
|
|
"loss": 6.249,
|
|
"mean_token_accuracy": 0.1511934891343117,
|
|
"num_tokens": 5143119.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.4086737632751465,
|
|
"epoch": 2.6146195922426654,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004649737832379534,
|
|
"loss": 6.0523,
|
|
"mean_token_accuracy": 0.17252246364951135,
|
|
"num_tokens": 5152167.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.614079427719116,
|
|
"epoch": 2.6195922426653406,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004647642410287057,
|
|
"loss": 6.3369,
|
|
"mean_token_accuracy": 0.15388641506433487,
|
|
"num_tokens": 5161962.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.604783010482788,
|
|
"epoch": 2.624564893088016,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004645541271343696,
|
|
"loss": 6.1713,
|
|
"mean_token_accuracy": 0.15520114451646805,
|
|
"num_tokens": 5171637.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.474964570999146,
|
|
"epoch": 2.629537543510691,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046434344218793566,
|
|
"loss": 6.1649,
|
|
"mean_token_accuracy": 0.15841156542301177,
|
|
"num_tokens": 5181839.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.563134241104126,
|
|
"epoch": 2.6345101939333664,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046413218682411485,
|
|
"loss": 6.2235,
|
|
"mean_token_accuracy": 0.15705921351909638,
|
|
"num_tokens": 5192237.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.641672277450562,
|
|
"epoch": 2.6394828443560416,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004639203616793365,
|
|
"loss": 6.3822,
|
|
"mean_token_accuracy": 0.14698525220155717,
|
|
"num_tokens": 5203879.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.636945486068726,
|
|
"epoch": 2.6444554947787173,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046370796739174657,
|
|
"loss": 6.2523,
|
|
"mean_token_accuracy": 0.14517115727066993,
|
|
"num_tokens": 5213456.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.578554010391235,
|
|
"epoch": 2.6494281452013926,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046349500460120556,
|
|
"loss": 6.1717,
|
|
"mean_token_accuracy": 0.1561637282371521,
|
|
"num_tokens": 5222234.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.56922755241394,
|
|
"epoch": 2.654400795624068,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046328147394928684,
|
|
"loss": 6.2047,
|
|
"mean_token_accuracy": 0.15863224118947983,
|
|
"num_tokens": 5230989.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.588573741912842,
|
|
"epoch": 2.659373446046743,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004630673760792741,
|
|
"loss": 6.2357,
|
|
"mean_token_accuracy": 0.15131169110536574,
|
|
"num_tokens": 5240607.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.641214084625244,
|
|
"epoch": 2.6643460964694183,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004628527116361603,
|
|
"loss": 6.3469,
|
|
"mean_token_accuracy": 0.14372803270816803,
|
|
"num_tokens": 5250837.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.614591264724732,
|
|
"epoch": 2.6693187468920936,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004626374812666449,
|
|
"loss": 6.2547,
|
|
"mean_token_accuracy": 0.15559754520654678,
|
|
"num_tokens": 5260542.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.551952123641968,
|
|
"epoch": 2.674291397314769,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046242168561913256,
|
|
"loss": 6.1382,
|
|
"mean_token_accuracy": 0.15862776190042496,
|
|
"num_tokens": 5272327.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.577802133560181,
|
|
"epoch": 2.679264047737444,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046220532534373067,
|
|
"loss": 6.2507,
|
|
"mean_token_accuracy": 0.15700989663600923,
|
|
"num_tokens": 5281407.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.673555850982666,
|
|
"epoch": 2.6842366981601193,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004619884010922477,
|
|
"loss": 6.2314,
|
|
"mean_token_accuracy": 0.14551012963056564,
|
|
"num_tokens": 5291617.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.652135515213013,
|
|
"epoch": 2.6892093485827946,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004617709135181913,
|
|
"loss": 6.2869,
|
|
"mean_token_accuracy": 0.1508165940642357,
|
|
"num_tokens": 5301174.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.612260627746582,
|
|
"epoch": 2.69418199900547,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046155286327676594,
|
|
"loss": 6.2634,
|
|
"mean_token_accuracy": 0.1465717986226082,
|
|
"num_tokens": 5311211.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.567706346511841,
|
|
"epoch": 2.699154649428145,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046133425102487134,
|
|
"loss": 6.2775,
|
|
"mean_token_accuracy": 0.15217909440398217,
|
|
"num_tokens": 5321343.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.541842842102051,
|
|
"epoch": 2.7041272998508203,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046111507742110035,
|
|
"loss": 6.226,
|
|
"mean_token_accuracy": 0.16114234030246735,
|
|
"num_tokens": 5330849.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.589290046691895,
|
|
"epoch": 2.7090999502734956,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000460895343125737,
|
|
"loss": 6.3274,
|
|
"mean_token_accuracy": 0.14705052301287652,
|
|
"num_tokens": 5340836.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.620804357528686,
|
|
"epoch": 2.714072600696171,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046067504880075425,
|
|
"loss": 6.196,
|
|
"mean_token_accuracy": 0.15342168286442756,
|
|
"num_tokens": 5351093.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.5411396503448485,
|
|
"epoch": 2.719045251118846,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004604541951098122,
|
|
"loss": 6.232,
|
|
"mean_token_accuracy": 0.15440142005681992,
|
|
"num_tokens": 5360859.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.480919265747071,
|
|
"epoch": 2.724017901541522,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004602327827182565,
|
|
"loss": 6.0145,
|
|
"mean_token_accuracy": 0.16869030743837357,
|
|
"num_tokens": 5370591.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.544408130645752,
|
|
"epoch": 2.728990551964197,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004600108122931156,
|
|
"loss": 6.1408,
|
|
"mean_token_accuracy": 0.16343188434839248,
|
|
"num_tokens": 5380019.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.521484470367431,
|
|
"epoch": 2.7339632023868723,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045978828450309907,
|
|
"loss": 6.2292,
|
|
"mean_token_accuracy": 0.14990266114473344,
|
|
"num_tokens": 5389558.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.584447431564331,
|
|
"epoch": 2.7389358528095475,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004595652000185958,
|
|
"loss": 6.2601,
|
|
"mean_token_accuracy": 0.1561106637120247,
|
|
"num_tokens": 5398305.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.64121265411377,
|
|
"epoch": 2.743908503232223,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045934155951167163,
|
|
"loss": 6.3307,
|
|
"mean_token_accuracy": 0.1506441690027714,
|
|
"num_tokens": 5406886.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.585583114624024,
|
|
"epoch": 2.748881153654898,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004591173636560675,
|
|
"loss": 6.3164,
|
|
"mean_token_accuracy": 0.15496177673339845,
|
|
"num_tokens": 5416404.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.596722555160523,
|
|
"epoch": 2.7538538040775733,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004588926131271974,
|
|
"loss": 6.2029,
|
|
"mean_token_accuracy": 0.15936118364334106,
|
|
"num_tokens": 5425746.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.602889490127564,
|
|
"epoch": 2.7588264545002485,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045866730860214655,
|
|
"loss": 6.2849,
|
|
"mean_token_accuracy": 0.15213748216629028,
|
|
"num_tokens": 5436089.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.511632776260376,
|
|
"epoch": 2.763799104922924,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004584414507596687,
|
|
"loss": 6.2099,
|
|
"mean_token_accuracy": 0.15665143132209777,
|
|
"num_tokens": 5446270.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.593560409545899,
|
|
"epoch": 2.768771755345599,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045821504028018494,
|
|
"loss": 6.226,
|
|
"mean_token_accuracy": 0.15678894370794297,
|
|
"num_tokens": 5455476.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.589935922622681,
|
|
"epoch": 2.7737444057682747,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004579880778457811,
|
|
"loss": 6.2353,
|
|
"mean_token_accuracy": 0.14878517910838127,
|
|
"num_tokens": 5463813.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.544997453689575,
|
|
"epoch": 2.77871705619095,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004577605641402056,
|
|
"loss": 6.2309,
|
|
"mean_token_accuracy": 0.15037454515695572,
|
|
"num_tokens": 5473127.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.48923134803772,
|
|
"epoch": 2.7836897066136252,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004575324998488681,
|
|
"loss": 6.1759,
|
|
"mean_token_accuracy": 0.15556782856583595,
|
|
"num_tokens": 5481595.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.502620124816895,
|
|
"epoch": 2.7886623570363005,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004573038856588364,
|
|
"loss": 6.2408,
|
|
"mean_token_accuracy": 0.15178192779421806,
|
|
"num_tokens": 5491495.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.537503099441528,
|
|
"epoch": 2.7936350074589757,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004570747222588354,
|
|
"loss": 6.2194,
|
|
"mean_token_accuracy": 0.15515500903129578,
|
|
"num_tokens": 5500702.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.581478118896484,
|
|
"epoch": 2.798607657881651,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045684501033924443,
|
|
"loss": 6.1427,
|
|
"mean_token_accuracy": 0.16013326719403267,
|
|
"num_tokens": 5510223.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.623900842666626,
|
|
"epoch": 2.8035803083043263,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045661475059209503,
|
|
"loss": 6.3313,
|
|
"mean_token_accuracy": 0.14360361844301223,
|
|
"num_tokens": 5522063.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.576891708374023,
|
|
"epoch": 2.8085529587270015,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004563839437110695,
|
|
"loss": 6.1952,
|
|
"mean_token_accuracy": 0.1603342965245247,
|
|
"num_tokens": 5531323.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.553138637542725,
|
|
"epoch": 2.8135256091496768,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045615259039149826,
|
|
"loss": 6.2064,
|
|
"mean_token_accuracy": 0.15693445652723312,
|
|
"num_tokens": 5540637.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.549783372879029,
|
|
"epoch": 2.818498259572352,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045592069133035783,
|
|
"loss": 6.1219,
|
|
"mean_token_accuracy": 0.1656826376914978,
|
|
"num_tokens": 5550992.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.546471977233887,
|
|
"epoch": 2.8234709099950273,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004556882472262691,
|
|
"loss": 6.1844,
|
|
"mean_token_accuracy": 0.1544899344444275,
|
|
"num_tokens": 5561140.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.549879169464111,
|
|
"epoch": 2.8284435604177025,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004554552587794948,
|
|
"loss": 6.2056,
|
|
"mean_token_accuracy": 0.1543327383697033,
|
|
"num_tokens": 5570457.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.550867891311645,
|
|
"epoch": 2.8334162108403778,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004552217266919375,
|
|
"loss": 6.3728,
|
|
"mean_token_accuracy": 0.1408549502491951,
|
|
"num_tokens": 5579950.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.595098209381104,
|
|
"epoch": 2.838388861263053,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004549876516671376,
|
|
"loss": 6.2553,
|
|
"mean_token_accuracy": 0.15306972712278366,
|
|
"num_tokens": 5589441.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.5484644889831545,
|
|
"epoch": 2.8433615116857283,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045475303441027114,
|
|
"loss": 6.0986,
|
|
"mean_token_accuracy": 0.16014811620116234,
|
|
"num_tokens": 5598583.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.528734111785889,
|
|
"epoch": 2.8483341621084035,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004545178756281479,
|
|
"loss": 6.1706,
|
|
"mean_token_accuracy": 0.15799631029367447,
|
|
"num_tokens": 5609198.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.578134822845459,
|
|
"epoch": 2.853306812531079,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00045428217602920876,
|
|
"loss": 6.1921,
|
|
"mean_token_accuracy": 0.1512961983680725,
|
|
"num_tokens": 5620425.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.454615831375122,
|
|
"epoch": 2.8582794629537545,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045404593632352385,
|
|
"loss": 6.0212,
|
|
"mean_token_accuracy": 0.17213779538869858,
|
|
"num_tokens": 5629578.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.545614957809448,
|
|
"epoch": 2.8632521133764297,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004538091572227909,
|
|
"loss": 6.2252,
|
|
"mean_token_accuracy": 0.14892149716615677,
|
|
"num_tokens": 5639817.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.655675983428955,
|
|
"epoch": 2.868224763799105,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004535718394403319,
|
|
"loss": 6.3275,
|
|
"mean_token_accuracy": 0.15373293459415435,
|
|
"num_tokens": 5649130.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.584187078475952,
|
|
"epoch": 2.87319741422178,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004533339836910924,
|
|
"loss": 6.2432,
|
|
"mean_token_accuracy": 0.15679741352796556,
|
|
"num_tokens": 5659672.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.529476976394653,
|
|
"epoch": 2.8781700646444555,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004530955906916382,
|
|
"loss": 6.2914,
|
|
"mean_token_accuracy": 0.15016578435897826,
|
|
"num_tokens": 5669491.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.418554973602295,
|
|
"epoch": 2.8831427150671307,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004528566611601536,
|
|
"loss": 6.0044,
|
|
"mean_token_accuracy": 0.173332779109478,
|
|
"num_tokens": 5679444.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.603994369506836,
|
|
"epoch": 2.888115365489806,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004526171958164396,
|
|
"loss": 6.3383,
|
|
"mean_token_accuracy": 0.14239170923829078,
|
|
"num_tokens": 5690997.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.576897382736206,
|
|
"epoch": 2.893088015912481,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045237719538191107,
|
|
"loss": 6.2259,
|
|
"mean_token_accuracy": 0.1546081230044365,
|
|
"num_tokens": 5700727.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.612170934677124,
|
|
"epoch": 2.898060666335157,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000452136660579595,
|
|
"loss": 6.2039,
|
|
"mean_token_accuracy": 0.15638039112091065,
|
|
"num_tokens": 5710809.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.385060024261475,
|
|
"epoch": 2.903033316757832,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045189559213412825,
|
|
"loss": 6.1048,
|
|
"mean_token_accuracy": 0.17112839221954346,
|
|
"num_tokens": 5720286.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.509190368652344,
|
|
"epoch": 2.9080059671805074,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004516539907717553,
|
|
"loss": 6.1945,
|
|
"mean_token_accuracy": 0.15884986966848375,
|
|
"num_tokens": 5730627.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.500278902053833,
|
|
"epoch": 2.9129786176031827,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045141185722032617,
|
|
"loss": 6.1087,
|
|
"mean_token_accuracy": 0.16135235279798507,
|
|
"num_tokens": 5739752.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.5710491180419925,
|
|
"epoch": 2.917951268025858,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045116919220929407,
|
|
"loss": 6.1841,
|
|
"mean_token_accuracy": 0.15795492380857468,
|
|
"num_tokens": 5748277.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.426823663711548,
|
|
"epoch": 2.922923918448533,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004509259964697133,
|
|
"loss": 6.0071,
|
|
"mean_token_accuracy": 0.17231556549668312,
|
|
"num_tokens": 5758632.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.489618110656738,
|
|
"epoch": 2.9278965688712084,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000450682270734237,
|
|
"loss": 6.165,
|
|
"mean_token_accuracy": 0.15990507751703262,
|
|
"num_tokens": 5769107.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.556905603408813,
|
|
"epoch": 2.9328692192938837,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004504380157371152,
|
|
"loss": 6.2055,
|
|
"mean_token_accuracy": 0.1508503556251526,
|
|
"num_tokens": 5779051.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.532180452346802,
|
|
"epoch": 2.937841869716559,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045019323221419204,
|
|
"loss": 6.1775,
|
|
"mean_token_accuracy": 0.15258841216564178,
|
|
"num_tokens": 5788826.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.501386976242065,
|
|
"epoch": 2.942814520139234,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044994792090290425,
|
|
"loss": 6.2019,
|
|
"mean_token_accuracy": 0.1528385914862156,
|
|
"num_tokens": 5798003.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.521900844573975,
|
|
"epoch": 2.9477871705619094,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044970208254227836,
|
|
"loss": 6.1786,
|
|
"mean_token_accuracy": 0.15506985485553743,
|
|
"num_tokens": 5807215.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.539266681671142,
|
|
"epoch": 2.9527598209845847,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044945571787292866,
|
|
"loss": 6.1803,
|
|
"mean_token_accuracy": 0.15545019134879112,
|
|
"num_tokens": 5816629.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.529702138900757,
|
|
"epoch": 2.95773247140726,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004492088276370552,
|
|
"loss": 6.1602,
|
|
"mean_token_accuracy": 0.1558448627591133,
|
|
"num_tokens": 5825840.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.630377101898193,
|
|
"epoch": 2.962705121829935,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004489614125784411,
|
|
"loss": 6.3459,
|
|
"mean_token_accuracy": 0.1436647929251194,
|
|
"num_tokens": 5835754.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.44536337852478,
|
|
"epoch": 2.9676777722526104,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004487134734424509,
|
|
"loss": 6.218,
|
|
"mean_token_accuracy": 0.15457739531993867,
|
|
"num_tokens": 5846062.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.61094036102295,
|
|
"epoch": 2.9726504226752857,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004484650109760276,
|
|
"loss": 6.2836,
|
|
"mean_token_accuracy": 0.1522247761487961,
|
|
"num_tokens": 5855589.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.59827880859375,
|
|
"epoch": 2.9776230730979614,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004482160259276911,
|
|
"loss": 6.0745,
|
|
"mean_token_accuracy": 0.16684725731611252,
|
|
"num_tokens": 5864917.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.528877544403076,
|
|
"epoch": 2.9825957235206366,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004479665190475355,
|
|
"loss": 6.1958,
|
|
"mean_token_accuracy": 0.15861882865428925,
|
|
"num_tokens": 5875530.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.9825957235206366,
|
|
"eval_entropy": 6.332148998014389,
|
|
"eval_loss": 6.598383903503418,
|
|
"eval_mean_token_accuracy": 0.13804876746841566,
|
|
"eval_num_tokens": 5875530.0,
|
|
"eval_runtime": 1.9357,
|
|
"eval_samples_per_second": 1790.058,
|
|
"eval_steps_per_second": 224.209,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.558258438110352,
|
|
"epoch": 2.987568373943312,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044771649108722695,
|
|
"loss": 6.2638,
|
|
"mean_token_accuracy": 0.15484196990728377,
|
|
"num_tokens": 5885363.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.613479280471802,
|
|
"epoch": 2.992541024365987,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004474659428000014,
|
|
"loss": 6.2599,
|
|
"mean_token_accuracy": 0.15506193190813064,
|
|
"num_tokens": 5895318.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.547708654403687,
|
|
"epoch": 2.9975136747886624,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044721487494066254,
|
|
"loss": 6.287,
|
|
"mean_token_accuracy": 0.15789175257086754,
|
|
"num_tokens": 5905742.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.528264628516303,
|
|
"epoch": 3.00198906016907,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000446963288265579,
|
|
"loss": 6.0331,
|
|
"mean_token_accuracy": 0.1738815555969874,
|
|
"num_tokens": 5913844.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.53761854171753,
|
|
"epoch": 3.0069617105917454,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044671118353268257,
|
|
"loss": 5.8994,
|
|
"mean_token_accuracy": 0.16046555265784263,
|
|
"num_tokens": 5923724.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.477243232727051,
|
|
"epoch": 3.0119343610144207,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044645856150146593,
|
|
"loss": 5.9257,
|
|
"mean_token_accuracy": 0.16766104996204376,
|
|
"num_tokens": 5933787.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.5848949432373045,
|
|
"epoch": 3.016907011437096,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044620542293297974,
|
|
"loss": 5.921,
|
|
"mean_token_accuracy": 0.16418616026639937,
|
|
"num_tokens": 5944186.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.543403625488281,
|
|
"epoch": 3.021879661859771,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004459517685898312,
|
|
"loss": 5.9213,
|
|
"mean_token_accuracy": 0.16770756840705872,
|
|
"num_tokens": 5954465.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.6111236095428465,
|
|
"epoch": 3.0268523122824464,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044569759923618113,
|
|
"loss": 5.9507,
|
|
"mean_token_accuracy": 0.16335516944527625,
|
|
"num_tokens": 5965047.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.439112806320191,
|
|
"epoch": 3.0318249627051217,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004454429156377419,
|
|
"loss": 5.7713,
|
|
"mean_token_accuracy": 0.1770850419998169,
|
|
"num_tokens": 5974855.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.4012593746185305,
|
|
"epoch": 3.036797613127797,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004451877185617751,
|
|
"loss": 5.7647,
|
|
"mean_token_accuracy": 0.17236735671758652,
|
|
"num_tokens": 5985965.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.5960510730743405,
|
|
"epoch": 3.041770263550472,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004449320087770894,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.16500016450881957,
|
|
"num_tokens": 5996646.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.511561489105224,
|
|
"epoch": 3.046742913973148,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004446757870540376,
|
|
"loss": 5.7872,
|
|
"mean_token_accuracy": 0.18356228917837142,
|
|
"num_tokens": 6005825.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.385194873809814,
|
|
"epoch": 3.051715564395823,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004444190541645153,
|
|
"loss": 5.7254,
|
|
"mean_token_accuracy": 0.17865629494190216,
|
|
"num_tokens": 6016909.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.501277923583984,
|
|
"epoch": 3.0566882148184984,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044416181088195776,
|
|
"loss": 5.9073,
|
|
"mean_token_accuracy": 0.16628172993659973,
|
|
"num_tokens": 6025511.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.466328859329224,
|
|
"epoch": 3.0616608652411736,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004439040579813379,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.17258121371269225,
|
|
"num_tokens": 6035114.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.5011450290679935,
|
|
"epoch": 3.066633515663849,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044364579623916385,
|
|
"loss": 5.8687,
|
|
"mean_token_accuracy": 0.16890793442726135,
|
|
"num_tokens": 6044963.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.377497243881225,
|
|
"epoch": 3.071606166086524,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004433870264334767,
|
|
"loss": 5.7456,
|
|
"mean_token_accuracy": 0.1834716647863388,
|
|
"num_tokens": 6054660.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.457064771652222,
|
|
"epoch": 3.0765788165091994,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004431277493438484,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.16877402365207672,
|
|
"num_tokens": 6064288.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.455753326416016,
|
|
"epoch": 3.0815514669318746,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004428679657513789,
|
|
"loss": 5.8414,
|
|
"mean_token_accuracy": 0.16629945784807204,
|
|
"num_tokens": 6074133.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.462636947631836,
|
|
"epoch": 3.08652411735455,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004426076764386939,
|
|
"loss": 5.8722,
|
|
"mean_token_accuracy": 0.1700928896665573,
|
|
"num_tokens": 6084314.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.387950229644775,
|
|
"epoch": 3.091496767777225,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044234688218994293,
|
|
"loss": 5.8313,
|
|
"mean_token_accuracy": 0.1730772614479065,
|
|
"num_tokens": 6094616.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.468109273910523,
|
|
"epoch": 3.0964694181999004,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004420855837907967,
|
|
"loss": 5.8518,
|
|
"mean_token_accuracy": 0.16851246505975723,
|
|
"num_tokens": 6104167.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.297282457351685,
|
|
"epoch": 3.1014420686225757,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044182378202844427,
|
|
"loss": 5.6372,
|
|
"mean_token_accuracy": 0.19177404791116714,
|
|
"num_tokens": 6113970.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.4837359428405765,
|
|
"epoch": 3.106414719045251,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044156147769159183,
|
|
"loss": 5.847,
|
|
"mean_token_accuracy": 0.1649303212761879,
|
|
"num_tokens": 6123482.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.41380124092102,
|
|
"epoch": 3.1113873694679266,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004412986715704591,
|
|
"loss": 5.8093,
|
|
"mean_token_accuracy": 0.17527061551809311,
|
|
"num_tokens": 6133682.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.416181898117065,
|
|
"epoch": 3.116360019890602,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004410353644567777,
|
|
"loss": 5.8446,
|
|
"mean_token_accuracy": 0.17112570852041245,
|
|
"num_tokens": 6143876.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.439123439788818,
|
|
"epoch": 3.121332670313277,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044077155714378836,
|
|
"loss": 5.9424,
|
|
"mean_token_accuracy": 0.1692077711224556,
|
|
"num_tokens": 6153615.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.448490142822266,
|
|
"epoch": 3.1263053207359524,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004405072504262391,
|
|
"loss": 5.8375,
|
|
"mean_token_accuracy": 0.17944991141557692,
|
|
"num_tokens": 6163572.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.39255690574646,
|
|
"epoch": 3.1312779711586276,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000440242445100382,
|
|
"loss": 5.8141,
|
|
"mean_token_accuracy": 0.17316634207963943,
|
|
"num_tokens": 6173085.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.41436357498169,
|
|
"epoch": 3.136250621581303,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043997714196397143,
|
|
"loss": 5.8794,
|
|
"mean_token_accuracy": 0.1649853803217411,
|
|
"num_tokens": 6183124.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.528604984283447,
|
|
"epoch": 3.141223272003978,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043971134181626173,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.16683800816535949,
|
|
"num_tokens": 6192817.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.4087806224823,
|
|
"epoch": 3.1461959224266534,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00043944504545800396,
|
|
"loss": 5.889,
|
|
"mean_token_accuracy": 0.17443156093358994,
|
|
"num_tokens": 6203643.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.448987293243408,
|
|
"epoch": 3.1511685728493286,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004391782536914447,
|
|
"loss": 5.8869,
|
|
"mean_token_accuracy": 0.1699537992477417,
|
|
"num_tokens": 6213712.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.415402555465699,
|
|
"epoch": 3.156141223272004,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043891096732032257,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.164980186522007,
|
|
"num_tokens": 6223962.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.428661155700683,
|
|
"epoch": 3.161113873694679,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043864318714986625,
|
|
"loss": 5.9534,
|
|
"mean_token_accuracy": 0.1640808552503586,
|
|
"num_tokens": 6235273.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.427452564239502,
|
|
"epoch": 3.1660865241173544,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043837491398679234,
|
|
"loss": 5.7539,
|
|
"mean_token_accuracy": 0.17510842382907868,
|
|
"num_tokens": 6245161.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.3870316505432125,
|
|
"epoch": 3.17105917454003,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004381061486393025,
|
|
"loss": 5.8337,
|
|
"mean_token_accuracy": 0.17053545266389847,
|
|
"num_tokens": 6253300.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.349837112426758,
|
|
"epoch": 3.1760318249627053,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004378368919170809,
|
|
"loss": 5.859,
|
|
"mean_token_accuracy": 0.17130498588085175,
|
|
"num_tokens": 6263525.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.386288833618164,
|
|
"epoch": 3.1810044753853806,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043756714463129244,
|
|
"loss": 5.8383,
|
|
"mean_token_accuracy": 0.1708736464381218,
|
|
"num_tokens": 6274010.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.420946311950684,
|
|
"epoch": 3.185977125808056,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004372969075945796,
|
|
"loss": 5.8914,
|
|
"mean_token_accuracy": 0.17023267447948456,
|
|
"num_tokens": 6283442.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.377214527130127,
|
|
"epoch": 3.190949776230731,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004370261816210604,
|
|
"loss": 5.8448,
|
|
"mean_token_accuracy": 0.17260374277830123,
|
|
"num_tokens": 6293498.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.30924391746521,
|
|
"epoch": 3.1959224266534063,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004367549675263258,
|
|
"loss": 5.6774,
|
|
"mean_token_accuracy": 0.1980217769742012,
|
|
"num_tokens": 6303557.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.497784233093261,
|
|
"epoch": 3.2008950770760816,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004364832661274373,
|
|
"loss": 6.0248,
|
|
"mean_token_accuracy": 0.16211080476641654,
|
|
"num_tokens": 6314304.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.478355836868286,
|
|
"epoch": 3.205867727498757,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00043621107824292444,
|
|
"loss": 5.8778,
|
|
"mean_token_accuracy": 0.1674222782254219,
|
|
"num_tokens": 6324959.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.27369270324707,
|
|
"epoch": 3.210840377921432,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004359384046927824,
|
|
"loss": 5.6701,
|
|
"mean_token_accuracy": 0.19252893775701524,
|
|
"num_tokens": 6335909.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.436891555786133,
|
|
"epoch": 3.2158130283441073,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004356652462984693,
|
|
"loss": 5.9221,
|
|
"mean_token_accuracy": 0.16505683660507203,
|
|
"num_tokens": 6346659.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.372613477706909,
|
|
"epoch": 3.2207856787667826,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043539160388290416,
|
|
"loss": 5.7054,
|
|
"mean_token_accuracy": 0.1911502107977867,
|
|
"num_tokens": 6357264.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.444086313247681,
|
|
"epoch": 3.225758329189458,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004351174782704641,
|
|
"loss": 5.9432,
|
|
"mean_token_accuracy": 0.16422172337770463,
|
|
"num_tokens": 6366107.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.312821531295777,
|
|
"epoch": 3.230730979612133,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043484287028698176,
|
|
"loss": 5.7501,
|
|
"mean_token_accuracy": 0.18167751729488374,
|
|
"num_tokens": 6375505.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.4700196266174315,
|
|
"epoch": 3.2357036300348083,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043456778075974306,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.1607461929321289,
|
|
"num_tokens": 6384701.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.462629222869873,
|
|
"epoch": 3.240676280457484,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043429221051748487,
|
|
"loss": 5.7956,
|
|
"mean_token_accuracy": 0.17599567770957947,
|
|
"num_tokens": 6393707.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.380648612976074,
|
|
"epoch": 3.2456489308801593,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043401616039039174,
|
|
"loss": 5.9002,
|
|
"mean_token_accuracy": 0.17014744579792024,
|
|
"num_tokens": 6404277.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.422473192214966,
|
|
"epoch": 3.2506215813028345,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004337396312100944,
|
|
"loss": 5.949,
|
|
"mean_token_accuracy": 0.16356757581233977,
|
|
"num_tokens": 6415275.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.418755340576172,
|
|
"epoch": 3.2555942317255098,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004334626238096665,
|
|
"loss": 5.8483,
|
|
"mean_token_accuracy": 0.1670474588871002,
|
|
"num_tokens": 6425598.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.375892972946167,
|
|
"epoch": 3.260566882148185,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004331851390236227,
|
|
"loss": 5.8049,
|
|
"mean_token_accuracy": 0.17047521620988845,
|
|
"num_tokens": 6435685.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.3282698631286625,
|
|
"epoch": 3.2655395325708603,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043290717768791533,
|
|
"loss": 5.8136,
|
|
"mean_token_accuracy": 0.17749550342559814,
|
|
"num_tokens": 6444950.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.36887001991272,
|
|
"epoch": 3.2705121829935355,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043262874063993285,
|
|
"loss": 5.8141,
|
|
"mean_token_accuracy": 0.17236246913671494,
|
|
"num_tokens": 6454066.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.399455833435058,
|
|
"epoch": 3.2754848334162108,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004323498287184967,
|
|
"loss": 5.9489,
|
|
"mean_token_accuracy": 0.16807875037193298,
|
|
"num_tokens": 6464912.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.42661566734314,
|
|
"epoch": 3.280457483838886,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043207044276385884,
|
|
"loss": 5.9739,
|
|
"mean_token_accuracy": 0.16375386714935303,
|
|
"num_tokens": 6474178.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.313139295578003,
|
|
"epoch": 3.2854301342615613,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043179058361769924,
|
|
"loss": 5.6697,
|
|
"mean_token_accuracy": 0.19066467136144638,
|
|
"num_tokens": 6485703.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.4194597721099855,
|
|
"epoch": 3.2904027846842365,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004315102521231237,
|
|
"loss": 5.8091,
|
|
"mean_token_accuracy": 0.16903553307056426,
|
|
"num_tokens": 6495309.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.412261343002319,
|
|
"epoch": 3.295375435106912,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004312294491246609,
|
|
"loss": 5.77,
|
|
"mean_token_accuracy": 0.17215309143066407,
|
|
"num_tokens": 6505498.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.326479959487915,
|
|
"epoch": 3.3003480855295875,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004309481754682597,
|
|
"loss": 5.806,
|
|
"mean_token_accuracy": 0.1742389440536499,
|
|
"num_tokens": 6514656.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.402297782897949,
|
|
"epoch": 3.3053207359522627,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043066643200128724,
|
|
"loss": 5.8725,
|
|
"mean_token_accuracy": 0.1691297024488449,
|
|
"num_tokens": 6524549.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.347615909576416,
|
|
"epoch": 3.310293386374938,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004303842195725259,
|
|
"loss": 5.7772,
|
|
"mean_token_accuracy": 0.179095321893692,
|
|
"num_tokens": 6533861.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.463089370727539,
|
|
"epoch": 3.3152660367976132,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043010153903217074,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.16502568125724792,
|
|
"num_tokens": 6543676.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.460142040252686,
|
|
"epoch": 3.3202386872202885,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004298183912318273,
|
|
"loss": 5.8604,
|
|
"mean_token_accuracy": 0.17189764529466628,
|
|
"num_tokens": 6552263.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.312333631515503,
|
|
"epoch": 3.3252113376429637,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042953477702450845,
|
|
"loss": 5.8152,
|
|
"mean_token_accuracy": 0.17552696019411088,
|
|
"num_tokens": 6560960.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.280674886703491,
|
|
"epoch": 3.330183988065639,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004292506972646324,
|
|
"loss": 5.7198,
|
|
"mean_token_accuracy": 0.18778837472200394,
|
|
"num_tokens": 6572345.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.393889236450195,
|
|
"epoch": 3.3351566384883142,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042896615280802006,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.1674686163663864,
|
|
"num_tokens": 6582274.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.353819513320923,
|
|
"epoch": 3.3401292889109895,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042868114451189164,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.17429812699556352,
|
|
"num_tokens": 6592486.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.305846548080444,
|
|
"epoch": 3.3451019393336647,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004283956732348655,
|
|
"loss": 5.7481,
|
|
"mean_token_accuracy": 0.1889159560203552,
|
|
"num_tokens": 6600876.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.365818738937378,
|
|
"epoch": 3.35007458975634,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042810973983695426,
|
|
"loss": 5.7754,
|
|
"mean_token_accuracy": 0.17992894053459169,
|
|
"num_tokens": 6610679.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.40848298072815,
|
|
"epoch": 3.3550472401790152,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004278233451795629,
|
|
"loss": 5.8956,
|
|
"mean_token_accuracy": 0.16785149723291398,
|
|
"num_tokens": 6620779.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.432049369812011,
|
|
"epoch": 3.3600198906016905,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000427536490125486,
|
|
"loss": 5.8394,
|
|
"mean_token_accuracy": 0.17895749509334563,
|
|
"num_tokens": 6630807.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.293106746673584,
|
|
"epoch": 3.364992541024366,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042724917553890507,
|
|
"loss": 5.7367,
|
|
"mean_token_accuracy": 0.18019191771745682,
|
|
"num_tokens": 6640530.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.336058807373047,
|
|
"epoch": 3.3699651914470414,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042696140228538616,
|
|
"loss": 5.8026,
|
|
"mean_token_accuracy": 0.17796725630760193,
|
|
"num_tokens": 6650845.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.352558946609497,
|
|
"epoch": 3.3749378418697167,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042667317123187696,
|
|
"loss": 5.8354,
|
|
"mean_token_accuracy": 0.16683292388916016,
|
|
"num_tokens": 6660455.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.341485786437988,
|
|
"epoch": 3.379910492292392,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042638448324670424,
|
|
"loss": 5.8051,
|
|
"mean_token_accuracy": 0.17608758062124252,
|
|
"num_tokens": 6670493.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.364583444595337,
|
|
"epoch": 3.384883142715067,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004260953391995716,
|
|
"loss": 5.9278,
|
|
"mean_token_accuracy": 0.16586780101060866,
|
|
"num_tokens": 6680598.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.223124837875366,
|
|
"epoch": 3.3898557931377424,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00042580573996155646,
|
|
"loss": 5.6784,
|
|
"mean_token_accuracy": 0.18944381773471833,
|
|
"num_tokens": 6691569.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.443700122833252,
|
|
"epoch": 3.3948284435604177,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004255156864051074,
|
|
"loss": 5.9601,
|
|
"mean_token_accuracy": 0.16494829207658768,
|
|
"num_tokens": 6702570.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.259357929229736,
|
|
"epoch": 3.399801093983093,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042522517940404186,
|
|
"loss": 5.6453,
|
|
"mean_token_accuracy": 0.19756894260644914,
|
|
"num_tokens": 6711628.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.4040148735046385,
|
|
"epoch": 3.404773744405768,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004249342198335433,
|
|
"loss": 5.8481,
|
|
"mean_token_accuracy": 0.16746716350317,
|
|
"num_tokens": 6721828.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.2174993515014645,
|
|
"epoch": 3.4097463948284434,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042464280857015856,
|
|
"loss": 5.6989,
|
|
"mean_token_accuracy": 0.18602155297994613,
|
|
"num_tokens": 6731903.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.246883201599121,
|
|
"epoch": 3.4147190452511187,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042435094649179525,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.17920604199171067,
|
|
"num_tokens": 6743236.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.257451009750366,
|
|
"epoch": 3.419691695673794,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004240586344777191,
|
|
"loss": 5.6828,
|
|
"mean_token_accuracy": 0.1838308408856392,
|
|
"num_tokens": 6753291.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.467526388168335,
|
|
"epoch": 3.4246643460964696,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004237658734085514,
|
|
"loss": 5.8711,
|
|
"mean_token_accuracy": 0.16685103327035905,
|
|
"num_tokens": 6762635.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.391610431671142,
|
|
"epoch": 3.429636996519145,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042347266416626624,
|
|
"loss": 5.8371,
|
|
"mean_token_accuracy": 0.17143007814884187,
|
|
"num_tokens": 6772101.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.329255390167236,
|
|
"epoch": 3.43460964694182,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004231790076341877,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.17412067353725433,
|
|
"num_tokens": 6781563.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.356528663635254,
|
|
"epoch": 3.4395822973644954,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004228849046969876,
|
|
"loss": 5.8218,
|
|
"mean_token_accuracy": 0.17599937021732331,
|
|
"num_tokens": 6790589.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.380900859832764,
|
|
"epoch": 3.4445549477871706,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004225903562406826,
|
|
"loss": 5.8409,
|
|
"mean_token_accuracy": 0.1737615644931793,
|
|
"num_tokens": 6799842.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.380544948577881,
|
|
"epoch": 3.449527598209846,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004222953631526313,
|
|
"loss": 5.8868,
|
|
"mean_token_accuracy": 0.17071893215179443,
|
|
"num_tokens": 6809166.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.373358201980591,
|
|
"epoch": 3.454500248632521,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000421999926321532,
|
|
"loss": 5.9006,
|
|
"mean_token_accuracy": 0.16730448007583618,
|
|
"num_tokens": 6818255.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.364234399795532,
|
|
"epoch": 3.4594728990551964,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004217040466374199,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.17515202909708022,
|
|
"num_tokens": 6828628.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.375469112396241,
|
|
"epoch": 3.4644455494778716,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042140772499166403,
|
|
"loss": 5.922,
|
|
"mean_token_accuracy": 0.16836985275149347,
|
|
"num_tokens": 6839053.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.4062180519104,
|
|
"epoch": 3.469418199900547,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004211109622769652,
|
|
"loss": 5.8689,
|
|
"mean_token_accuracy": 0.17099950909614564,
|
|
"num_tokens": 6849226.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.335232257843018,
|
|
"epoch": 3.474390850323222,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042081375938735283,
|
|
"loss": 5.8227,
|
|
"mean_token_accuracy": 0.17957468330860138,
|
|
"num_tokens": 6858867.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.366913080215454,
|
|
"epoch": 3.4793635007458974,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042051611721818233,
|
|
"loss": 5.8601,
|
|
"mean_token_accuracy": 0.1743657797574997,
|
|
"num_tokens": 6869171.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.404802656173706,
|
|
"epoch": 3.4843361511685726,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042021803666613273,
|
|
"loss": 5.8258,
|
|
"mean_token_accuracy": 0.1717052936553955,
|
|
"num_tokens": 6878582.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.2995504379272464,
|
|
"epoch": 3.489308801591248,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041991951862920365,
|
|
"loss": 5.8891,
|
|
"mean_token_accuracy": 0.16955186575651168,
|
|
"num_tokens": 6887522.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.286713123321533,
|
|
"epoch": 3.4942814520139236,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004196205640067124,
|
|
"loss": 5.7849,
|
|
"mean_token_accuracy": 0.177962289750576,
|
|
"num_tokens": 6897556.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.384421348571777,
|
|
"epoch": 3.499254102436599,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000419321173699292,
|
|
"loss": 5.841,
|
|
"mean_token_accuracy": 0.1733379766345024,
|
|
"num_tokens": 6908219.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.39391303062439,
|
|
"epoch": 3.504226752859274,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004190213486088877,
|
|
"loss": 5.8653,
|
|
"mean_token_accuracy": 0.1665293201804161,
|
|
"num_tokens": 6918175.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.379923391342163,
|
|
"epoch": 3.5091994032819493,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004187210896387546,
|
|
"loss": 5.9885,
|
|
"mean_token_accuracy": 0.17161076962947847,
|
|
"num_tokens": 6927076.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.385812044143677,
|
|
"epoch": 3.5141720537046246,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004184203976934552,
|
|
"loss": 5.8709,
|
|
"mean_token_accuracy": 0.17569606006145477,
|
|
"num_tokens": 6936568.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.287307024002075,
|
|
"epoch": 3.5191447041273,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00041811927367885605,
|
|
"loss": 5.8348,
|
|
"mean_token_accuracy": 0.17747514545917512,
|
|
"num_tokens": 6945494.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.389292001724243,
|
|
"epoch": 3.524117354549975,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004178177185021254,
|
|
"loss": 5.813,
|
|
"mean_token_accuracy": 0.17755813002586365,
|
|
"num_tokens": 6955397.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.400311517715454,
|
|
"epoch": 3.5290900049726504,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004175157330717305,
|
|
"loss": 5.9413,
|
|
"mean_token_accuracy": 0.16256057173013688,
|
|
"num_tokens": 6965431.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.296528291702271,
|
|
"epoch": 3.5340626553953256,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000417213318297435,
|
|
"loss": 5.7933,
|
|
"mean_token_accuracy": 0.17628138661384582,
|
|
"num_tokens": 6975457.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.311970138549805,
|
|
"epoch": 3.539035305818001,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041691047509029555,
|
|
"loss": 5.6917,
|
|
"mean_token_accuracy": 0.18531340062618257,
|
|
"num_tokens": 6984691.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.232669401168823,
|
|
"epoch": 3.5440079562406765,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041660720436265987,
|
|
"loss": 5.724,
|
|
"mean_token_accuracy": 0.18893563747406006,
|
|
"num_tokens": 6994052.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.337771034240722,
|
|
"epoch": 3.548980606663352,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004163035070281634,
|
|
"loss": 5.8485,
|
|
"mean_token_accuracy": 0.17254940718412398,
|
|
"num_tokens": 7003625.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.338563537597656,
|
|
"epoch": 3.553953257086027,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041599938400172686,
|
|
"loss": 5.9068,
|
|
"mean_token_accuracy": 0.1673935428261757,
|
|
"num_tokens": 7011967.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.39595103263855,
|
|
"epoch": 3.5589259075087023,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041569483619955354,
|
|
"loss": 5.9294,
|
|
"mean_token_accuracy": 0.16915967166423798,
|
|
"num_tokens": 7021722.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.474276685714722,
|
|
"epoch": 3.5638985579313776,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041538986453912616,
|
|
"loss": 5.9505,
|
|
"mean_token_accuracy": 0.16677639335393907,
|
|
"num_tokens": 7031720.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.3342526912689205,
|
|
"epoch": 3.568871208354053,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004150844699392045,
|
|
"loss": 5.9137,
|
|
"mean_token_accuracy": 0.16426487863063813,
|
|
"num_tokens": 7040575.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.452836751937866,
|
|
"epoch": 3.573843858776728,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041477865331982264,
|
|
"loss": 5.8904,
|
|
"mean_token_accuracy": 0.16770058870315552,
|
|
"num_tokens": 7050035.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.253588581085205,
|
|
"epoch": 3.5788165091994033,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004144724156022857,
|
|
"loss": 5.6965,
|
|
"mean_token_accuracy": 0.18944824486970901,
|
|
"num_tokens": 7060950.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.315596294403076,
|
|
"epoch": 3.5837891596220786,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041416575770916763,
|
|
"loss": 5.8664,
|
|
"mean_token_accuracy": 0.16883061379194259,
|
|
"num_tokens": 7070384.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.448754215240479,
|
|
"epoch": 3.588761810044754,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041385868056430827,
|
|
"loss": 5.9713,
|
|
"mean_token_accuracy": 0.1580202743411064,
|
|
"num_tokens": 7081118.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.414922142028809,
|
|
"epoch": 3.593734460467429,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004135511850928102,
|
|
"loss": 5.9207,
|
|
"mean_token_accuracy": 0.16742333620786667,
|
|
"num_tokens": 7091157.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.332775831222534,
|
|
"epoch": 3.5987071108901043,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00041324327222103655,
|
|
"loss": 5.9013,
|
|
"mean_token_accuracy": 0.167673459649086,
|
|
"num_tokens": 7101005.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.307641124725341,
|
|
"epoch": 3.6036797613127796,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041293494287660785,
|
|
"loss": 5.8559,
|
|
"mean_token_accuracy": 0.16871901601552963,
|
|
"num_tokens": 7109944.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.449390554428101,
|
|
"epoch": 3.608652411735455,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041262619798839914,
|
|
"loss": 5.9227,
|
|
"mean_token_accuracy": 0.16109448969364165,
|
|
"num_tokens": 7119073.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.391678667068481,
|
|
"epoch": 3.61362506215813,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041231703848653746,
|
|
"loss": 5.9224,
|
|
"mean_token_accuracy": 0.1656919687986374,
|
|
"num_tokens": 7128920.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.3840882778167725,
|
|
"epoch": 3.6185977125808053,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041200746530239903,
|
|
"loss": 5.9078,
|
|
"mean_token_accuracy": 0.16942816078662873,
|
|
"num_tokens": 7138691.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.371076679229736,
|
|
"epoch": 3.623570363003481,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004116974793686061,
|
|
"loss": 5.828,
|
|
"mean_token_accuracy": 0.17209259867668153,
|
|
"num_tokens": 7148362.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.320016956329345,
|
|
"epoch": 3.6285430134261563,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004113870816190244,
|
|
"loss": 5.8487,
|
|
"mean_token_accuracy": 0.1747833862900734,
|
|
"num_tokens": 7157138.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.367502164840698,
|
|
"epoch": 3.6335156638488315,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004110762729887605,
|
|
"loss": 5.9621,
|
|
"mean_token_accuracy": 0.16389233320951463,
|
|
"num_tokens": 7167425.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.35754132270813,
|
|
"epoch": 3.6384883142715068,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041076505441415873,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.16941047012805938,
|
|
"num_tokens": 7177059.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.379943132400513,
|
|
"epoch": 3.643460964694182,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004104534268327982,
|
|
"loss": 5.9079,
|
|
"mean_token_accuracy": 0.16824225932359696,
|
|
"num_tokens": 7186435.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.339653587341308,
|
|
"epoch": 3.6484336151168573,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004101413911834904,
|
|
"loss": 5.8966,
|
|
"mean_token_accuracy": 0.16970329284667968,
|
|
"num_tokens": 7197248.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.348167276382446,
|
|
"epoch": 3.6534062655395325,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040982894840627614,
|
|
"loss": 5.8673,
|
|
"mean_token_accuracy": 0.16948134452104568,
|
|
"num_tokens": 7206042.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.396376466751098,
|
|
"epoch": 3.6583789159622078,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004095160994424228,
|
|
"loss": 5.9174,
|
|
"mean_token_accuracy": 0.1717466950416565,
|
|
"num_tokens": 7215822.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.338345909118653,
|
|
"epoch": 3.663351566384883,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040920284523442125,
|
|
"loss": 5.9175,
|
|
"mean_token_accuracy": 0.16754038333892823,
|
|
"num_tokens": 7226131.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.308998537063599,
|
|
"epoch": 3.6683242168075583,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040888918672598337,
|
|
"loss": 5.83,
|
|
"mean_token_accuracy": 0.17459045946598054,
|
|
"num_tokens": 7234095.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.3149699687957765,
|
|
"epoch": 3.673296867230234,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000408575124862039,
|
|
"loss": 5.7837,
|
|
"mean_token_accuracy": 0.1854248061776161,
|
|
"num_tokens": 7243184.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.37895188331604,
|
|
"epoch": 3.678269517652909,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004082606605887331,
|
|
"loss": 5.8824,
|
|
"mean_token_accuracy": 0.16198230236768724,
|
|
"num_tokens": 7253654.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.307571363449097,
|
|
"epoch": 3.6832421680755845,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040794579485342304,
|
|
"loss": 5.7516,
|
|
"mean_token_accuracy": 0.16902582496404647,
|
|
"num_tokens": 7262063.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.236040210723877,
|
|
"epoch": 3.6882148184982597,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00040763052860467547,
|
|
"loss": 5.7638,
|
|
"mean_token_accuracy": 0.18761724531650542,
|
|
"num_tokens": 7272490.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.359106397628784,
|
|
"epoch": 3.693187468920935,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004073148627922637,
|
|
"loss": 5.9772,
|
|
"mean_token_accuracy": 0.16209677755832672,
|
|
"num_tokens": 7282938.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.386785411834717,
|
|
"epoch": 3.69816011934361,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004069987983671648,
|
|
"loss": 5.8348,
|
|
"mean_token_accuracy": 0.16977072209119798,
|
|
"num_tokens": 7293486.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.266835451126099,
|
|
"epoch": 3.7031327697662855,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040668233628155664,
|
|
"loss": 5.8261,
|
|
"mean_token_accuracy": 0.16591209620237352,
|
|
"num_tokens": 7302806.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.302120018005371,
|
|
"epoch": 3.7081054201889607,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004063654774888153,
|
|
"loss": 5.9029,
|
|
"mean_token_accuracy": 0.16786832064390184,
|
|
"num_tokens": 7312250.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.452527475357056,
|
|
"epoch": 3.713078070611636,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004060482229435116,
|
|
"loss": 5.8361,
|
|
"mean_token_accuracy": 0.17729832530021666,
|
|
"num_tokens": 7321512.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.309455299377442,
|
|
"epoch": 3.7180507210343112,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004057305736014089,
|
|
"loss": 5.8754,
|
|
"mean_token_accuracy": 0.1790863022208214,
|
|
"num_tokens": 7331467.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.277251052856445,
|
|
"epoch": 3.7230233714569865,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004054125304194598,
|
|
"loss": 5.837,
|
|
"mean_token_accuracy": 0.1708943650126457,
|
|
"num_tokens": 7340064.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.312506055831909,
|
|
"epoch": 3.7279960218796617,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040509409435580355,
|
|
"loss": 5.8363,
|
|
"mean_token_accuracy": 0.177268148958683,
|
|
"num_tokens": 7350138.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.3638567447662355,
|
|
"epoch": 3.732968672302337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004047752663697627,
|
|
"loss": 5.8877,
|
|
"mean_token_accuracy": 0.17120849788188935,
|
|
"num_tokens": 7359613.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.343510627746582,
|
|
"epoch": 3.7379413227250122,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004044560474218409,
|
|
"loss": 5.8319,
|
|
"mean_token_accuracy": 0.17689885199069977,
|
|
"num_tokens": 7368754.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.357437086105347,
|
|
"epoch": 3.7429139731476875,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040413643847371923,
|
|
"loss": 5.8074,
|
|
"mean_token_accuracy": 0.1733761414885521,
|
|
"num_tokens": 7378930.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.408187961578369,
|
|
"epoch": 3.7478866235703627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004038164404882539,
|
|
"loss": 5.8985,
|
|
"mean_token_accuracy": 0.16141457259654998,
|
|
"num_tokens": 7387281.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.340773391723633,
|
|
"epoch": 3.7528592739930384,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004034960544294731,
|
|
"loss": 5.775,
|
|
"mean_token_accuracy": 0.17530253231525422,
|
|
"num_tokens": 7397039.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.291833782196045,
|
|
"epoch": 3.7578319244157137,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040317528126257413,
|
|
"loss": 5.8559,
|
|
"mean_token_accuracy": 0.17591804265975952,
|
|
"num_tokens": 7406771.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.333402013778686,
|
|
"epoch": 3.762804574838389,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004028541219539205,
|
|
"loss": 5.7934,
|
|
"mean_token_accuracy": 0.18379120975732804,
|
|
"num_tokens": 7415743.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.173971033096313,
|
|
"epoch": 3.767777225261064,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004025325774710391,
|
|
"loss": 5.6106,
|
|
"mean_token_accuracy": 0.1862991198897362,
|
|
"num_tokens": 7424243.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.2883594036102295,
|
|
"epoch": 3.7727498756837394,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040221064878261684,
|
|
"loss": 5.9115,
|
|
"mean_token_accuracy": 0.16776654422283171,
|
|
"num_tokens": 7434273.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.366536808013916,
|
|
"epoch": 3.7777225261064147,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004018883368584985,
|
|
"loss": 5.8505,
|
|
"mean_token_accuracy": 0.17147254794836045,
|
|
"num_tokens": 7444640.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.253841876983643,
|
|
"epoch": 3.78269517652909,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 0.00040156564266968336,
|
|
"loss": 5.7069,
|
|
"mean_token_accuracy": 0.17850466668605805,
|
|
"num_tokens": 7453774.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.332209348678589,
|
|
"epoch": 3.787667826951765,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040124256718832214,
|
|
"loss": 5.8365,
|
|
"mean_token_accuracy": 0.17283500880002975,
|
|
"num_tokens": 7464288.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.367890071868897,
|
|
"epoch": 3.7926404773744404,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004009191113877141,
|
|
"loss": 5.8986,
|
|
"mean_token_accuracy": 0.16678115129470825,
|
|
"num_tokens": 7474626.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.3670775413513185,
|
|
"epoch": 3.797613127797116,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004005952762423048,
|
|
"loss": 5.8827,
|
|
"mean_token_accuracy": 0.18107935041189194,
|
|
"num_tokens": 7483393.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.36843204498291,
|
|
"epoch": 3.8025857782197914,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004002710627276821,
|
|
"loss": 5.9141,
|
|
"mean_token_accuracy": 0.1663992002606392,
|
|
"num_tokens": 7492197.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.286423873901367,
|
|
"epoch": 3.8075584286424666,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003999464718205739,
|
|
"loss": 5.8659,
|
|
"mean_token_accuracy": 0.1762652188539505,
|
|
"num_tokens": 7501822.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.236625337600708,
|
|
"epoch": 3.812531079065142,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003996215044988451,
|
|
"loss": 5.7723,
|
|
"mean_token_accuracy": 0.18264362812042237,
|
|
"num_tokens": 7512686.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.391349744796753,
|
|
"epoch": 3.817503729487817,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039929616174149454,
|
|
"loss": 5.8625,
|
|
"mean_token_accuracy": 0.1673120528459549,
|
|
"num_tokens": 7522593.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.367191410064697,
|
|
"epoch": 3.8224763799104924,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039897044452865207,
|
|
"loss": 5.9293,
|
|
"mean_token_accuracy": 0.16810093447566032,
|
|
"num_tokens": 7532448.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.200035333633423,
|
|
"epoch": 3.8274490303331676,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003986443538415756,
|
|
"loss": 5.6588,
|
|
"mean_token_accuracy": 0.19849494844675064,
|
|
"num_tokens": 7543008.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.281786870956421,
|
|
"epoch": 3.832421680755843,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003983178906626484,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.17977537512779235,
|
|
"num_tokens": 7553236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.359131240844727,
|
|
"epoch": 3.837394331178518,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039799105597537564,
|
|
"loss": 5.8923,
|
|
"mean_token_accuracy": 0.17551862299442292,
|
|
"num_tokens": 7561583.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.295608329772949,
|
|
"epoch": 3.8423669816011934,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003976638507643818,
|
|
"loss": 5.8741,
|
|
"mean_token_accuracy": 0.16841931194067,
|
|
"num_tokens": 7571290.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.30658483505249,
|
|
"epoch": 3.8473396320238686,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003973362760154076,
|
|
"loss": 5.8791,
|
|
"mean_token_accuracy": 0.16741953343153,
|
|
"num_tokens": 7579976.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.322652816772461,
|
|
"epoch": 3.852312282446544,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003970083327153071,
|
|
"loss": 5.8371,
|
|
"mean_token_accuracy": 0.17004714012145997,
|
|
"num_tokens": 7589796.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.3199516296386715,
|
|
"epoch": 3.857284932869219,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003966800218520445,
|
|
"loss": 5.7762,
|
|
"mean_token_accuracy": 0.1815151035785675,
|
|
"num_tokens": 7598976.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.3197808265686035,
|
|
"epoch": 3.8622575832918944,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003963513444146914,
|
|
"loss": 5.9455,
|
|
"mean_token_accuracy": 0.16175675690174102,
|
|
"num_tokens": 7607955.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.288658380508423,
|
|
"epoch": 3.8672302337145696,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003960223013934239,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.17342263460159302,
|
|
"num_tokens": 7619822.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.2453389167785645,
|
|
"epoch": 3.872202884137245,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003956928937795192,
|
|
"loss": 5.7298,
|
|
"mean_token_accuracy": 0.1853141352534294,
|
|
"num_tokens": 7630481.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.318514966964722,
|
|
"epoch": 3.8771755345599206,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000395363122565353,
|
|
"loss": 5.8578,
|
|
"mean_token_accuracy": 0.17130321711301805,
|
|
"num_tokens": 7640312.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.463041543960571,
|
|
"epoch": 3.882148184982596,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039503298874439655,
|
|
"loss": 6.1069,
|
|
"mean_token_accuracy": 0.151725222915411,
|
|
"num_tokens": 7649792.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.3688465595245365,
|
|
"epoch": 3.887120835405271,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003947024933112131,
|
|
"loss": 5.8208,
|
|
"mean_token_accuracy": 0.17764586061239243,
|
|
"num_tokens": 7659458.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.32335352897644,
|
|
"epoch": 3.8920934858279463,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003943716372614556,
|
|
"loss": 5.8429,
|
|
"mean_token_accuracy": 0.17711288183927537,
|
|
"num_tokens": 7668770.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.221032524108887,
|
|
"epoch": 3.8970661362506216,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003940404215918634,
|
|
"loss": 5.7589,
|
|
"mean_token_accuracy": 0.17949200719594954,
|
|
"num_tokens": 7677067.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.231208372116089,
|
|
"epoch": 3.902038786673297,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039370884730025897,
|
|
"loss": 5.8273,
|
|
"mean_token_accuracy": 0.17425628006458282,
|
|
"num_tokens": 7687723.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.403788328170776,
|
|
"epoch": 3.907011437095972,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003933769153855455,
|
|
"loss": 5.9759,
|
|
"mean_token_accuracy": 0.167946694791317,
|
|
"num_tokens": 7696706.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.357872200012207,
|
|
"epoch": 3.9119840875186473,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003930446268477034,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.171970397233963,
|
|
"num_tokens": 7706356.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.331662940979004,
|
|
"epoch": 3.9169567379413226,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003927119826877875,
|
|
"loss": 5.9,
|
|
"mean_token_accuracy": 0.17476721107959747,
|
|
"num_tokens": 7717255.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.286929416656494,
|
|
"epoch": 3.921929388363998,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003923789839079238,
|
|
"loss": 5.9007,
|
|
"mean_token_accuracy": 0.16835979968309403,
|
|
"num_tokens": 7728493.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.362729930877686,
|
|
"epoch": 3.9269020387866735,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039204563151130683,
|
|
"loss": 5.8464,
|
|
"mean_token_accuracy": 0.16873070299625398,
|
|
"num_tokens": 7738582.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.367378902435303,
|
|
"epoch": 3.931874689209349,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003917119265021965,
|
|
"loss": 5.9358,
|
|
"mean_token_accuracy": 0.17097740918397902,
|
|
"num_tokens": 7748486.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.228780937194824,
|
|
"epoch": 3.936847339632024,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003913778698859146,
|
|
"loss": 5.6533,
|
|
"mean_token_accuracy": 0.18755111545324327,
|
|
"num_tokens": 7758033.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.287197113037109,
|
|
"epoch": 3.9418199900546993,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003910434626688427,
|
|
"loss": 5.7426,
|
|
"mean_token_accuracy": 0.17604231387376784,
|
|
"num_tokens": 7768577.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.3055624008178714,
|
|
"epoch": 3.9467926404773745,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003907087058584183,
|
|
"loss": 5.8759,
|
|
"mean_token_accuracy": 0.16810473799705505,
|
|
"num_tokens": 7778546.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.306858491897583,
|
|
"epoch": 3.95176529090005,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039037360046313186,
|
|
"loss": 5.8714,
|
|
"mean_token_accuracy": 0.17229774892330169,
|
|
"num_tokens": 7787523.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.37854495048523,
|
|
"epoch": 3.956737941322725,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00039003814749252454,
|
|
"loss": 5.8416,
|
|
"mean_token_accuracy": 0.17057503014802933,
|
|
"num_tokens": 7798198.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.3898365020751955,
|
|
"epoch": 3.9617105917454003,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038970234795718424,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.15822484493255615,
|
|
"num_tokens": 7808137.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.319563722610473,
|
|
"epoch": 3.9666832421680756,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000389366202868743,
|
|
"loss": 5.8318,
|
|
"mean_token_accuracy": 0.17717629820108413,
|
|
"num_tokens": 7817367.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.171590900421142,
|
|
"epoch": 3.971655892590751,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003890297132398739,
|
|
"loss": 5.6651,
|
|
"mean_token_accuracy": 0.18896573781967163,
|
|
"num_tokens": 7826419.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.302308082580566,
|
|
"epoch": 3.976628543013426,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003886928800842879,
|
|
"loss": 5.8515,
|
|
"mean_token_accuracy": 0.17299700528383255,
|
|
"num_tokens": 7834984.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.976628543013426,
|
|
"eval_entropy": 6.1573500688174905,
|
|
"eval_loss": 6.485073566436768,
|
|
"eval_mean_token_accuracy": 0.14689584266102534,
|
|
"eval_num_tokens": 7834984.0,
|
|
"eval_runtime": 1.9247,
|
|
"eval_samples_per_second": 1800.271,
|
|
"eval_steps_per_second": 225.488,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 6.3073570728302,
|
|
"epoch": 3.9816011934361013,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038835570441673095,
|
|
"loss": 5.8286,
|
|
"mean_token_accuracy": 0.17251945286989212,
|
|
"num_tokens": 7844781.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 6.325926876068115,
|
|
"epoch": 3.9865738438587766,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003880181872529809,
|
|
"loss": 5.9022,
|
|
"mean_token_accuracy": 0.17272959500551224,
|
|
"num_tokens": 7854553.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 6.3461753845214846,
|
|
"epoch": 3.991546494281452,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003876803296098443,
|
|
"loss": 5.851,
|
|
"mean_token_accuracy": 0.1694219410419464,
|
|
"num_tokens": 7863900.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.3461230278015135,
|
|
"epoch": 3.996519144704127,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003873421325051535,
|
|
"loss": 5.8215,
|
|
"mean_token_accuracy": 0.17737256735563278,
|
|
"num_tokens": 7873008.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 6.313788202073839,
|
|
"epoch": 4.000994530084535,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038700359695776346,
|
|
"loss": 5.8122,
|
|
"mean_token_accuracy": 0.17067693008316887,
|
|
"num_tokens": 7882489.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.289868259429932,
|
|
"epoch": 4.00596718050721,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003866647239875487,
|
|
"loss": 5.6156,
|
|
"mean_token_accuracy": 0.17952480018138886,
|
|
"num_tokens": 7893370.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 6.319686079025269,
|
|
"epoch": 4.010939830929885,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038632551461540023,
|
|
"loss": 5.4996,
|
|
"mean_token_accuracy": 0.19372497946023942,
|
|
"num_tokens": 7903815.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 6.381075000762939,
|
|
"epoch": 4.015912481352561,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003859859698632228,
|
|
"loss": 5.622,
|
|
"mean_token_accuracy": 0.17955474257469178,
|
|
"num_tokens": 7913365.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.348626804351807,
|
|
"epoch": 4.020885131775236,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003856460907539312,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.1917821064591408,
|
|
"num_tokens": 7922973.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.383808946609497,
|
|
"epoch": 4.025857782197911,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003853058783114478,
|
|
"loss": 5.5508,
|
|
"mean_token_accuracy": 0.1891060560941696,
|
|
"num_tokens": 7932860.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 6.283610725402832,
|
|
"epoch": 4.030830432620586,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003849653335606987,
|
|
"loss": 5.5726,
|
|
"mean_token_accuracy": 0.18631523847579956,
|
|
"num_tokens": 7942524.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 6.371997356414795,
|
|
"epoch": 4.035803083043262,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003846244575276117,
|
|
"loss": 5.6502,
|
|
"mean_token_accuracy": 0.17859771251678466,
|
|
"num_tokens": 7952289.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.227728843688965,
|
|
"epoch": 4.040775733465938,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038428325123911205,
|
|
"loss": 5.5005,
|
|
"mean_token_accuracy": 0.18740858286619186,
|
|
"num_tokens": 7962176.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.271157073974609,
|
|
"epoch": 4.045748383888613,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003839417157231205,
|
|
"loss": 5.5822,
|
|
"mean_token_accuracy": 0.18633224666118622,
|
|
"num_tokens": 7971300.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 6.2161705017089846,
|
|
"epoch": 4.050721034311288,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003835998520085492,
|
|
"loss": 5.4947,
|
|
"mean_token_accuracy": 0.19686799198389054,
|
|
"num_tokens": 7981135.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 6.27214503288269,
|
|
"epoch": 4.0556936847339635,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003832576611252993,
|
|
"loss": 5.5198,
|
|
"mean_token_accuracy": 0.19364309757947923,
|
|
"num_tokens": 7991327.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 6.335424280166626,
|
|
"epoch": 4.060666335156639,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003829151441042572,
|
|
"loss": 5.5981,
|
|
"mean_token_accuracy": 0.18398454040288925,
|
|
"num_tokens": 8001318.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 6.181913614273071,
|
|
"epoch": 4.065638985579314,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038257230197729237,
|
|
"loss": 5.4,
|
|
"mean_token_accuracy": 0.20491259396076203,
|
|
"num_tokens": 8010964.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 6.1811745166778564,
|
|
"epoch": 4.070611636001989,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038222913577725325,
|
|
"loss": 5.5482,
|
|
"mean_token_accuracy": 0.19156203120946885,
|
|
"num_tokens": 8020894.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 6.272135448455811,
|
|
"epoch": 4.0755842864246645,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038188564653796487,
|
|
"loss": 5.5181,
|
|
"mean_token_accuracy": 0.19015952050685883,
|
|
"num_tokens": 8030635.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 6.27701587677002,
|
|
"epoch": 4.08055693684734,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003815418352942252,
|
|
"loss": 5.47,
|
|
"mean_token_accuracy": 0.19391969591379166,
|
|
"num_tokens": 8040626.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.987170648574829,
|
|
"epoch": 4.085529587270015,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003811977030818026,
|
|
"loss": 5.2091,
|
|
"mean_token_accuracy": 0.22891474664211273,
|
|
"num_tokens": 8051835.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 6.2427380084991455,
|
|
"epoch": 4.09050223769269,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000380853250937432,
|
|
"loss": 5.5593,
|
|
"mean_token_accuracy": 0.18835939317941666,
|
|
"num_tokens": 8061818.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 6.296293592453003,
|
|
"epoch": 4.0954748881153655,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038050847989881236,
|
|
"loss": 5.6414,
|
|
"mean_token_accuracy": 0.19034629315137863,
|
|
"num_tokens": 8071693.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 6.1796546459198,
|
|
"epoch": 4.100447538538041,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003801633910046035,
|
|
"loss": 5.4197,
|
|
"mean_token_accuracy": 0.20188747048377992,
|
|
"num_tokens": 8082304.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 6.268434143066406,
|
|
"epoch": 4.105420188960716,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003798179852944225,
|
|
"loss": 5.523,
|
|
"mean_token_accuracy": 0.187708979845047,
|
|
"num_tokens": 8092319.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 6.274722957611084,
|
|
"epoch": 4.110392839383391,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037947226380884115,
|
|
"loss": 5.529,
|
|
"mean_token_accuracy": 0.19465973675251008,
|
|
"num_tokens": 8100894.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 6.2412919998168945,
|
|
"epoch": 4.1153654898060665,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037912622758938243,
|
|
"loss": 5.6374,
|
|
"mean_token_accuracy": 0.17828676700592042,
|
|
"num_tokens": 8110029.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 6.2836192607879635,
|
|
"epoch": 4.120338140228742,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003787798776785174,
|
|
"loss": 5.5163,
|
|
"mean_token_accuracy": 0.18564724177122116,
|
|
"num_tokens": 8118715.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 6.237198066711426,
|
|
"epoch": 4.125310790651417,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003784332151196622,
|
|
"loss": 5.5155,
|
|
"mean_token_accuracy": 0.19903987497091294,
|
|
"num_tokens": 8128778.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 6.239213562011718,
|
|
"epoch": 4.130283441074092,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037808624095717503,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.1869506224989891,
|
|
"num_tokens": 8138126.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 6.236686134338379,
|
|
"epoch": 4.1352560914967675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003777389562363526,
|
|
"loss": 5.5321,
|
|
"mean_token_accuracy": 0.19594503790140153,
|
|
"num_tokens": 8147971.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 6.308943128585815,
|
|
"epoch": 4.140228741919443,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00037739136200342743,
|
|
"loss": 5.5943,
|
|
"mean_token_accuracy": 0.18684487491846086,
|
|
"num_tokens": 8157853.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 6.197887325286866,
|
|
"epoch": 4.145201392342118,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037704345930556426,
|
|
"loss": 5.5029,
|
|
"mean_token_accuracy": 0.19374612867832183,
|
|
"num_tokens": 8167187.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.179135656356811,
|
|
"epoch": 4.150174042764793,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037669524919085716,
|
|
"loss": 5.4657,
|
|
"mean_token_accuracy": 0.19783916622400283,
|
|
"num_tokens": 8175492.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.184837722778321,
|
|
"epoch": 4.1551466931874685,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003763467327083264,
|
|
"loss": 5.5633,
|
|
"mean_token_accuracy": 0.19013711661100388,
|
|
"num_tokens": 8185356.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 6.130315494537354,
|
|
"epoch": 4.160119343610144,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00037599791090791527,
|
|
"loss": 5.392,
|
|
"mean_token_accuracy": 0.20041110068559648,
|
|
"num_tokens": 8194833.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 6.3327052116394045,
|
|
"epoch": 4.165091994032819,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003756487848404868,
|
|
"loss": 5.6212,
|
|
"mean_token_accuracy": 0.18605939596891402,
|
|
"num_tokens": 8205040.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 6.311398887634278,
|
|
"epoch": 4.170064644455495,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003752993555578205,
|
|
"loss": 5.5892,
|
|
"mean_token_accuracy": 0.18565354794263839,
|
|
"num_tokens": 8214769.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 6.114333391189575,
|
|
"epoch": 4.17503729487817,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003749496241126095,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.20991736203432082,
|
|
"num_tokens": 8225007.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 6.305995655059815,
|
|
"epoch": 4.180009945300846,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00037459959155845726,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.18618288934230803,
|
|
"num_tokens": 8234543.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 6.206717491149902,
|
|
"epoch": 4.184982595723521,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003742492589498744,
|
|
"loss": 5.4648,
|
|
"mean_token_accuracy": 0.19633654057979583,
|
|
"num_tokens": 8243790.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 6.205511903762817,
|
|
"epoch": 4.189955246146196,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037389862734227525,
|
|
"loss": 5.5663,
|
|
"mean_token_accuracy": 0.18985246866941452,
|
|
"num_tokens": 8253144.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 6.242594623565674,
|
|
"epoch": 4.194927896568871,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037354769779197514,
|
|
"loss": 5.533,
|
|
"mean_token_accuracy": 0.19226989895105362,
|
|
"num_tokens": 8262782.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.232176828384399,
|
|
"epoch": 4.199900546991547,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037319647135618695,
|
|
"loss": 5.5573,
|
|
"mean_token_accuracy": 0.1875952884554863,
|
|
"num_tokens": 8271544.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 6.226108741760254,
|
|
"epoch": 4.204873197414222,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00037284494909301793,
|
|
"loss": 5.5314,
|
|
"mean_token_accuracy": 0.19108885675668716,
|
|
"num_tokens": 8280574.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 6.235661029815674,
|
|
"epoch": 4.209845847836897,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003724931320614663,
|
|
"loss": 5.5052,
|
|
"mean_token_accuracy": 0.19775112122297286,
|
|
"num_tokens": 8290318.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 6.220176935195923,
|
|
"epoch": 4.214818498259572,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037214102132141874,
|
|
"loss": 5.5329,
|
|
"mean_token_accuracy": 0.19377434402704238,
|
|
"num_tokens": 8299619.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 6.134536266326904,
|
|
"epoch": 4.219791148682248,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003717886179336465,
|
|
"loss": 5.493,
|
|
"mean_token_accuracy": 0.2016959384083748,
|
|
"num_tokens": 8310085.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 6.188360023498535,
|
|
"epoch": 4.224763799104923,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037143592295980243,
|
|
"loss": 5.5466,
|
|
"mean_token_accuracy": 0.19344982504844666,
|
|
"num_tokens": 8320673.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 6.2078221321105955,
|
|
"epoch": 4.229736449527598,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003710829374624181,
|
|
"loss": 5.5115,
|
|
"mean_token_accuracy": 0.19061054140329362,
|
|
"num_tokens": 8331073.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 6.263370704650879,
|
|
"epoch": 4.2347090999502734,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003707296625048999,
|
|
"loss": 5.6095,
|
|
"mean_token_accuracy": 0.1825783893465996,
|
|
"num_tokens": 8341313.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 6.263539409637451,
|
|
"epoch": 4.239681750372949,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003703760991515264,
|
|
"loss": 5.6136,
|
|
"mean_token_accuracy": 0.1801331400871277,
|
|
"num_tokens": 8351004.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 6.1020032405853275,
|
|
"epoch": 4.244654400795624,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00037002224846744526,
|
|
"loss": 5.3522,
|
|
"mean_token_accuracy": 0.20875437557697296,
|
|
"num_tokens": 8361083.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.238088178634643,
|
|
"epoch": 4.249627051218299,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036966811151866943,
|
|
"loss": 5.6353,
|
|
"mean_token_accuracy": 0.18680690675973893,
|
|
"num_tokens": 8372162.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 6.323531770706177,
|
|
"epoch": 4.2545997016409745,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036931368937207436,
|
|
"loss": 5.6252,
|
|
"mean_token_accuracy": 0.18430815935134887,
|
|
"num_tokens": 8382259.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 6.2864021301269535,
|
|
"epoch": 4.25957235206365,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003689589830953948,
|
|
"loss": 5.5716,
|
|
"mean_token_accuracy": 0.19354351460933686,
|
|
"num_tokens": 8392275.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 6.206379508972168,
|
|
"epoch": 4.264545002486325,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003686039937572213,
|
|
"loss": 5.5414,
|
|
"mean_token_accuracy": 0.19028273820877076,
|
|
"num_tokens": 8402481.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 6.1601286888122555,
|
|
"epoch": 4.269517652909,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003682487224269972,
|
|
"loss": 5.5624,
|
|
"mean_token_accuracy": 0.18609723299741746,
|
|
"num_tokens": 8412609.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 6.209527158737183,
|
|
"epoch": 4.2744903033316755,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003678931701750155,
|
|
"loss": 5.5997,
|
|
"mean_token_accuracy": 0.18573936969041824,
|
|
"num_tokens": 8421396.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 6.196327447891235,
|
|
"epoch": 4.279462953754351,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036753733807241536,
|
|
"loss": 5.548,
|
|
"mean_token_accuracy": 0.18889155685901643,
|
|
"num_tokens": 8430076.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 6.194756746292114,
|
|
"epoch": 4.284435604177026,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00036718122719117907,
|
|
"loss": 5.3865,
|
|
"mean_token_accuracy": 0.20747558474540712,
|
|
"num_tokens": 8440199.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 6.226442909240722,
|
|
"epoch": 4.289408254599701,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003668248386041288,
|
|
"loss": 5.5241,
|
|
"mean_token_accuracy": 0.19460933208465575,
|
|
"num_tokens": 8449463.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 6.200280618667603,
|
|
"epoch": 4.294380905022377,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00036646817338492334,
|
|
"loss": 5.4165,
|
|
"mean_token_accuracy": 0.19430915415287017,
|
|
"num_tokens": 8459824.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 6.26341290473938,
|
|
"epoch": 4.299353555445053,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003661112326080548,
|
|
"loss": 5.5642,
|
|
"mean_token_accuracy": 0.18831698596477509,
|
|
"num_tokens": 8468778.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 6.196838665008545,
|
|
"epoch": 4.304326205867728,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003657540173488455,
|
|
"loss": 5.5039,
|
|
"mean_token_accuracy": 0.19172408878803254,
|
|
"num_tokens": 8478542.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 6.095868444442749,
|
|
"epoch": 4.309298856290403,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003653965286834446,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.2042805403470993,
|
|
"num_tokens": 8489652.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 6.242799711227417,
|
|
"epoch": 4.314271506713078,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003650387676888251,
|
|
"loss": 5.597,
|
|
"mean_token_accuracy": 0.18471022546291352,
|
|
"num_tokens": 8499951.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 6.30907711982727,
|
|
"epoch": 4.319244157135754,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003646807354427802,
|
|
"loss": 5.7468,
|
|
"mean_token_accuracy": 0.17232859879732132,
|
|
"num_tokens": 8510664.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 6.30261721611023,
|
|
"epoch": 4.324216807558429,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036432243302392025,
|
|
"loss": 5.5882,
|
|
"mean_token_accuracy": 0.18933797627687454,
|
|
"num_tokens": 8519768.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 6.275367546081543,
|
|
"epoch": 4.329189457981104,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036396386151166975,
|
|
"loss": 5.6603,
|
|
"mean_token_accuracy": 0.18092515468597412,
|
|
"num_tokens": 8530909.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 6.2741623878479,
|
|
"epoch": 4.334162108403779,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003636050219862638,
|
|
"loss": 5.6587,
|
|
"mean_token_accuracy": 0.18164392411708832,
|
|
"num_tokens": 8541357.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 6.25959358215332,
|
|
"epoch": 4.339134758826455,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036324591552874493,
|
|
"loss": 5.575,
|
|
"mean_token_accuracy": 0.18306329101324081,
|
|
"num_tokens": 8550819.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 6.287986373901367,
|
|
"epoch": 4.34410740924913,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00036288654322095973,
|
|
"loss": 5.638,
|
|
"mean_token_accuracy": 0.18536092191934586,
|
|
"num_tokens": 8561577.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 6.3235283374786375,
|
|
"epoch": 4.349080059671805,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036252690614555557,
|
|
"loss": 5.6395,
|
|
"mean_token_accuracy": 0.17943083345890046,
|
|
"num_tokens": 8572590.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 6.240670490264892,
|
|
"epoch": 4.35405271009448,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003621670053859778,
|
|
"loss": 5.6197,
|
|
"mean_token_accuracy": 0.1840710684657097,
|
|
"num_tokens": 8583559.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 6.214067220687866,
|
|
"epoch": 4.359025360517156,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000361806842026466,
|
|
"loss": 5.6421,
|
|
"mean_token_accuracy": 0.1799193039536476,
|
|
"num_tokens": 8592845.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 6.054467725753784,
|
|
"epoch": 4.363998010939831,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036144641715205077,
|
|
"loss": 5.3594,
|
|
"mean_token_accuracy": 0.20440316647291185,
|
|
"num_tokens": 8602862.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 6.048689651489258,
|
|
"epoch": 4.368970661362506,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003610857318485507,
|
|
"loss": 5.4669,
|
|
"mean_token_accuracy": 0.2083234503865242,
|
|
"num_tokens": 8613793.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 6.092635440826416,
|
|
"epoch": 4.373943311785181,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00036072478720256885,
|
|
"loss": 5.3401,
|
|
"mean_token_accuracy": 0.2118215724825859,
|
|
"num_tokens": 8623086.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 6.062120628356934,
|
|
"epoch": 4.378915962207857,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003603635843014895,
|
|
"loss": 5.4217,
|
|
"mean_token_accuracy": 0.21053182780742646,
|
|
"num_tokens": 8632392.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 6.248952341079712,
|
|
"epoch": 4.383888612630532,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00036000212423347523,
|
|
"loss": 5.6613,
|
|
"mean_token_accuracy": 0.18171160519123078,
|
|
"num_tokens": 8642452.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 6.270582914352417,
|
|
"epoch": 4.388861263053207,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000359640408087463,
|
|
"loss": 5.6354,
|
|
"mean_token_accuracy": 0.18422549664974214,
|
|
"num_tokens": 8651431.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 6.310577917098999,
|
|
"epoch": 4.393833913475882,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003592784369531615,
|
|
"loss": 5.7063,
|
|
"mean_token_accuracy": 0.17800814360380174,
|
|
"num_tokens": 8661348.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 6.2487551212310795,
|
|
"epoch": 4.398806563898558,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035891621192104764,
|
|
"loss": 5.6256,
|
|
"mean_token_accuracy": 0.18288340121507646,
|
|
"num_tokens": 8671038.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 6.207523155212402,
|
|
"epoch": 4.403779214321233,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035855373408236294,
|
|
"loss": 5.5793,
|
|
"mean_token_accuracy": 0.1810170292854309,
|
|
"num_tokens": 8680115.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 6.255797719955444,
|
|
"epoch": 4.408751864743908,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003581910045291107,
|
|
"loss": 5.6375,
|
|
"mean_token_accuracy": 0.1825631782412529,
|
|
"num_tokens": 8689665.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 6.264122438430786,
|
|
"epoch": 4.413724515166583,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003578280243540526,
|
|
"loss": 5.605,
|
|
"mean_token_accuracy": 0.18357853293418885,
|
|
"num_tokens": 8700114.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 6.082082080841064,
|
|
"epoch": 4.4186971655892595,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035746479465070515,
|
|
"loss": 5.3579,
|
|
"mean_token_accuracy": 0.21364874243736268,
|
|
"num_tokens": 8710154.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 6.2371784210205075,
|
|
"epoch": 4.423669816011935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00035710131651333683,
|
|
"loss": 5.6899,
|
|
"mean_token_accuracy": 0.17978475987911224,
|
|
"num_tokens": 8720787.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 6.19954891204834,
|
|
"epoch": 4.42864246643461,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00035673759103696447,
|
|
"loss": 5.5457,
|
|
"mean_token_accuracy": 0.1913863867521286,
|
|
"num_tokens": 8731069.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 6.213446378707886,
|
|
"epoch": 4.433615116857285,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003563736193173499,
|
|
"loss": 5.5435,
|
|
"mean_token_accuracy": 0.1929771602153778,
|
|
"num_tokens": 8740149.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 6.330812644958496,
|
|
"epoch": 4.4385877672799605,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003560094024509969,
|
|
"loss": 5.6361,
|
|
"mean_token_accuracy": 0.18316797614097596,
|
|
"num_tokens": 8749622.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 6.17094988822937,
|
|
"epoch": 4.443560417702636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035564494153514777,
|
|
"loss": 5.5461,
|
|
"mean_token_accuracy": 0.1913030281662941,
|
|
"num_tokens": 8759521.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 6.180709934234619,
|
|
"epoch": 4.448533068125311,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035528023766778,
|
|
"loss": 5.6109,
|
|
"mean_token_accuracy": 0.19160889834165573,
|
|
"num_tokens": 8769387.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 6.262746810913086,
|
|
"epoch": 4.453505718547986,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003549152919476031,
|
|
"loss": 5.5806,
|
|
"mean_token_accuracy": 0.19440594762563707,
|
|
"num_tokens": 8777871.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 6.248330879211426,
|
|
"epoch": 4.4584783689706615,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035455010547405513,
|
|
"loss": 5.6404,
|
|
"mean_token_accuracy": 0.1835200846195221,
|
|
"num_tokens": 8787613.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 6.247484874725342,
|
|
"epoch": 4.463451019393337,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035418467934729934,
|
|
"loss": 5.5913,
|
|
"mean_token_accuracy": 0.17211652994155885,
|
|
"num_tokens": 8797075.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 6.206197357177734,
|
|
"epoch": 4.468423669816012,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035381901466822113,
|
|
"loss": 5.6572,
|
|
"mean_token_accuracy": 0.18328916877508164,
|
|
"num_tokens": 8806087.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 6.29120512008667,
|
|
"epoch": 4.473396320238687,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003534531125384244,
|
|
"loss": 5.6146,
|
|
"mean_token_accuracy": 0.18339093327522277,
|
|
"num_tokens": 8815358.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 6.2823561668396,
|
|
"epoch": 4.4783689706613625,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00035308697406022865,
|
|
"loss": 5.6404,
|
|
"mean_token_accuracy": 0.19202242493629457,
|
|
"num_tokens": 8825559.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 6.161588954925537,
|
|
"epoch": 4.483341621084038,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035272060033666503,
|
|
"loss": 5.5306,
|
|
"mean_token_accuracy": 0.1878397896885872,
|
|
"num_tokens": 8835136.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 6.192554330825805,
|
|
"epoch": 4.488314271506713,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003523539924714738,
|
|
"loss": 5.5759,
|
|
"mean_token_accuracy": 0.1929114744067192,
|
|
"num_tokens": 8843936.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 6.305029726028442,
|
|
"epoch": 4.493286921929388,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035198715156910016,
|
|
"loss": 5.7026,
|
|
"mean_token_accuracy": 0.17747835963964462,
|
|
"num_tokens": 8853990.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 6.184887313842774,
|
|
"epoch": 4.4982595723520635,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00035162007873469175,
|
|
"loss": 5.6474,
|
|
"mean_token_accuracy": 0.1849556103348732,
|
|
"num_tokens": 8863856.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 6.299717140197754,
|
|
"epoch": 4.503232222774739,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035125277507409483,
|
|
"loss": 5.6055,
|
|
"mean_token_accuracy": 0.18339322358369828,
|
|
"num_tokens": 8872846.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 6.108703136444092,
|
|
"epoch": 4.508204873197414,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00035088524169385075,
|
|
"loss": 5.4321,
|
|
"mean_token_accuracy": 0.20451173782348633,
|
|
"num_tokens": 8883323.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 6.235472106933594,
|
|
"epoch": 4.513177523620089,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035051747970119354,
|
|
"loss": 5.7069,
|
|
"mean_token_accuracy": 0.17567896693944932,
|
|
"num_tokens": 8893346.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 6.290061378479004,
|
|
"epoch": 4.5181501740427645,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035014949020404523,
|
|
"loss": 5.6155,
|
|
"mean_token_accuracy": 0.1845528244972229,
|
|
"num_tokens": 8902480.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 6.208347034454346,
|
|
"epoch": 4.52312282446544,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034978127431101365,
|
|
"loss": 5.5055,
|
|
"mean_token_accuracy": 0.19027355462312698,
|
|
"num_tokens": 8913496.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 6.2130719184875485,
|
|
"epoch": 4.528095474888115,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00034941283313138865,
|
|
"loss": 5.6404,
|
|
"mean_token_accuracy": 0.17837552726268768,
|
|
"num_tokens": 8924344.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 6.166843318939209,
|
|
"epoch": 4.53306812531079,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003490441677751387,
|
|
"loss": 5.5853,
|
|
"mean_token_accuracy": 0.18776468932628632,
|
|
"num_tokens": 8933602.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 6.279877662658691,
|
|
"epoch": 4.5380407757334655,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00034867527935290773,
|
|
"loss": 5.6063,
|
|
"mean_token_accuracy": 0.18621307015419006,
|
|
"num_tokens": 8943722.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 6.22838830947876,
|
|
"epoch": 4.543013426156142,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034830616897601146,
|
|
"loss": 5.5073,
|
|
"mean_token_accuracy": 0.19742414802312852,
|
|
"num_tokens": 8953535.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 6.1186288356781,
|
|
"epoch": 4.547986076578816,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003479368377564344,
|
|
"loss": 5.3854,
|
|
"mean_token_accuracy": 0.20221835374832153,
|
|
"num_tokens": 8963775.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 6.122174263000488,
|
|
"epoch": 4.552958727001492,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003475672868068265,
|
|
"loss": 5.4694,
|
|
"mean_token_accuracy": 0.19322815388441086,
|
|
"num_tokens": 8972822.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 6.138837242126465,
|
|
"epoch": 4.557931377424167,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034719751724049955,
|
|
"loss": 5.6318,
|
|
"mean_token_accuracy": 0.18484634160995483,
|
|
"num_tokens": 8983058.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 6.218491792678833,
|
|
"epoch": 4.562904027846843,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034682753017142387,
|
|
"loss": 5.6393,
|
|
"mean_token_accuracy": 0.18386669158935548,
|
|
"num_tokens": 8993069.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 6.2759607315063475,
|
|
"epoch": 4.567876678269518,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00034645732671422517,
|
|
"loss": 5.658,
|
|
"mean_token_accuracy": 0.18192542642354964,
|
|
"num_tokens": 9001471.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 6.253595399856567,
|
|
"epoch": 4.572849328692193,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003460869079841809,
|
|
"loss": 5.6147,
|
|
"mean_token_accuracy": 0.1831476002931595,
|
|
"num_tokens": 9012411.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 6.194453430175781,
|
|
"epoch": 4.577821979114868,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003457162750972172,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.195240581035614,
|
|
"num_tokens": 9024126.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 6.155144119262696,
|
|
"epoch": 4.582794629537544,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00034534542916990534,
|
|
"loss": 5.523,
|
|
"mean_token_accuracy": 0.19088055044412613,
|
|
"num_tokens": 9034355.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 6.204141092300415,
|
|
"epoch": 4.587767279960219,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034497437131945837,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.19011775702238082,
|
|
"num_tokens": 9043772.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 6.2302779197692875,
|
|
"epoch": 4.592739930382894,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003446031026637277,
|
|
"loss": 5.5631,
|
|
"mean_token_accuracy": 0.19415813386440278,
|
|
"num_tokens": 9053397.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 6.189073896408081,
|
|
"epoch": 4.597712580805569,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034423162432119995,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.18853379935026168,
|
|
"num_tokens": 9063904.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 6.208497714996338,
|
|
"epoch": 4.602685231228245,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003438599374109934,
|
|
"loss": 5.525,
|
|
"mean_token_accuracy": 0.19462729692459108,
|
|
"num_tokens": 9073146.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 6.068108367919922,
|
|
"epoch": 4.60765788165092,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034348804305285466,
|
|
"loss": 5.4066,
|
|
"mean_token_accuracy": 0.2108670249581337,
|
|
"num_tokens": 9083617.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 6.225018262863159,
|
|
"epoch": 4.612630532073595,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003431159423671552,
|
|
"loss": 5.5961,
|
|
"mean_token_accuracy": 0.18404778987169265,
|
|
"num_tokens": 9092639.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 6.198451662063599,
|
|
"epoch": 4.61760318249627,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034274363647488826,
|
|
"loss": 5.6737,
|
|
"mean_token_accuracy": 0.18323254734277725,
|
|
"num_tokens": 9103377.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 6.139889287948608,
|
|
"epoch": 4.622575832918946,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003423711264976652,
|
|
"loss": 5.484,
|
|
"mean_token_accuracy": 0.2019467458128929,
|
|
"num_tokens": 9112681.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 6.185818099975586,
|
|
"epoch": 4.627548483341621,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034199841355771226,
|
|
"loss": 5.527,
|
|
"mean_token_accuracy": 0.1966826096177101,
|
|
"num_tokens": 9121079.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 6.240311145782471,
|
|
"epoch": 4.632521133764296,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00034162549877786685,
|
|
"loss": 5.6153,
|
|
"mean_token_accuracy": 0.18978380262851716,
|
|
"num_tokens": 9131121.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 6.27166748046875,
|
|
"epoch": 4.6374937841869714,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003412523832815748,
|
|
"loss": 5.6468,
|
|
"mean_token_accuracy": 0.17824899703264235,
|
|
"num_tokens": 9140352.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 6.262094306945801,
|
|
"epoch": 4.642466434609647,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003408790681928865,
|
|
"loss": 5.6377,
|
|
"mean_token_accuracy": 0.17931518852710723,
|
|
"num_tokens": 9150892.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 6.185400342941284,
|
|
"epoch": 4.647439085032322,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003405055546364535,
|
|
"loss": 5.6542,
|
|
"mean_token_accuracy": 0.18153517991304396,
|
|
"num_tokens": 9160943.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 6.281888103485107,
|
|
"epoch": 4.652411735454997,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00034013184373752557,
|
|
"loss": 5.7123,
|
|
"mean_token_accuracy": 0.17921706438064575,
|
|
"num_tokens": 9171461.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 6.219541311264038,
|
|
"epoch": 4.6573843858776725,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003397579366219466,
|
|
"loss": 5.619,
|
|
"mean_token_accuracy": 0.18734892904758454,
|
|
"num_tokens": 9181991.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 6.224720239639282,
|
|
"epoch": 4.662357036300348,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0003393838344161519,
|
|
"loss": 5.4579,
|
|
"mean_token_accuracy": 0.19734470248222352,
|
|
"num_tokens": 9191715.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 6.207488250732422,
|
|
"epoch": 4.667329686723024,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033900953824716443,
|
|
"loss": 5.527,
|
|
"mean_token_accuracy": 0.19226354509592056,
|
|
"num_tokens": 9201338.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 6.199440622329712,
|
|
"epoch": 4.672302337145698,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003386350492425913,
|
|
"loss": 5.651,
|
|
"mean_token_accuracy": 0.18197681605815888,
|
|
"num_tokens": 9210970.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 6.241244602203369,
|
|
"epoch": 4.677274987568374,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033826036853062095,
|
|
"loss": 5.6133,
|
|
"mean_token_accuracy": 0.1823400780558586,
|
|
"num_tokens": 9221304.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 6.252964353561401,
|
|
"epoch": 4.68224763799105,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033788549724001897,
|
|
"loss": 5.5609,
|
|
"mean_token_accuracy": 0.18370507210493087,
|
|
"num_tokens": 9231386.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 6.236640739440918,
|
|
"epoch": 4.687220288413725,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033751043650012534,
|
|
"loss": 5.6507,
|
|
"mean_token_accuracy": 0.18542349934577942,
|
|
"num_tokens": 9240203.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 6.200273513793945,
|
|
"epoch": 4.6921929388364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003371351874408506,
|
|
"loss": 5.5852,
|
|
"mean_token_accuracy": 0.18651439547538756,
|
|
"num_tokens": 9249623.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 6.2684471130371096,
|
|
"epoch": 4.697165589259075,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003367597511926728,
|
|
"loss": 5.6648,
|
|
"mean_token_accuracy": 0.1781701147556305,
|
|
"num_tokens": 9259359.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 6.231116867065429,
|
|
"epoch": 4.702138239681751,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00033638412888663374,
|
|
"loss": 5.5179,
|
|
"mean_token_accuracy": 0.19265371263027192,
|
|
"num_tokens": 9267897.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 6.1611669063568115,
|
|
"epoch": 4.707110890104426,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003360083216543358,
|
|
"loss": 5.4862,
|
|
"mean_token_accuracy": 0.1967909574508667,
|
|
"num_tokens": 9277520.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 6.32455153465271,
|
|
"epoch": 4.712083540527101,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003356323306279387,
|
|
"loss": 5.7984,
|
|
"mean_token_accuracy": 0.16618154048919678,
|
|
"num_tokens": 9288157.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 6.220534896850586,
|
|
"epoch": 4.717056190949776,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033525615694015545,
|
|
"loss": 5.5416,
|
|
"mean_token_accuracy": 0.192324860394001,
|
|
"num_tokens": 9296698.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 6.262156677246094,
|
|
"epoch": 4.722028841372452,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00033487980172424955,
|
|
"loss": 5.6504,
|
|
"mean_token_accuracy": 0.18177213072776793,
|
|
"num_tokens": 9305880.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 6.267016792297364,
|
|
"epoch": 4.727001491795127,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033450326611403136,
|
|
"loss": 5.6729,
|
|
"mean_token_accuracy": 0.1822001278400421,
|
|
"num_tokens": 9316123.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 6.285989379882812,
|
|
"epoch": 4.731974142217802,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033412655124385494,
|
|
"loss": 5.6757,
|
|
"mean_token_accuracy": 0.18406474590301514,
|
|
"num_tokens": 9325914.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 6.092288398742676,
|
|
"epoch": 4.736946792640477,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003337496582486139,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.1954971134662628,
|
|
"num_tokens": 9336330.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 6.238794946670533,
|
|
"epoch": 4.741919443063153,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033337258826373896,
|
|
"loss": 5.6123,
|
|
"mean_token_accuracy": 0.17892003953456878,
|
|
"num_tokens": 9345358.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 6.240053367614746,
|
|
"epoch": 4.746892093485828,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033299534242519355,
|
|
"loss": 5.5487,
|
|
"mean_token_accuracy": 0.18934281319379806,
|
|
"num_tokens": 9355610.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 6.153910827636719,
|
|
"epoch": 4.751864743908503,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033261792186947133,
|
|
"loss": 5.6084,
|
|
"mean_token_accuracy": 0.19300020039081572,
|
|
"num_tokens": 9364450.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 6.211133003234863,
|
|
"epoch": 4.756837394331178,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033224032773359203,
|
|
"loss": 5.656,
|
|
"mean_token_accuracy": 0.18289080411195754,
|
|
"num_tokens": 9374426.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 6.1193889617919925,
|
|
"epoch": 4.761810044753854,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033186256115509845,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.20814799815416335,
|
|
"num_tokens": 9384067.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 6.131963872909546,
|
|
"epoch": 4.766782695176529,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003314846232720528,
|
|
"loss": 5.3672,
|
|
"mean_token_accuracy": 0.20634757578372956,
|
|
"num_tokens": 9394269.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 6.211862516403198,
|
|
"epoch": 4.771755345599204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033110651522303346,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.19461610317230224,
|
|
"num_tokens": 9403930.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 6.201242208480835,
|
|
"epoch": 4.776727996021879,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003307282381471314,
|
|
"loss": 5.5938,
|
|
"mean_token_accuracy": 0.17858741879463197,
|
|
"num_tokens": 9413520.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 6.1596307277679445,
|
|
"epoch": 4.781700646444555,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033034979318394676,
|
|
"loss": 5.5716,
|
|
"mean_token_accuracy": 0.188422554731369,
|
|
"num_tokens": 9423945.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 6.200018310546875,
|
|
"epoch": 4.78667329686723,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032997118147358554,
|
|
"loss": 5.6157,
|
|
"mean_token_accuracy": 0.19240077435970307,
|
|
"num_tokens": 9434139.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 6.226301050186157,
|
|
"epoch": 4.791645947289906,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000329592404156656,
|
|
"loss": 5.6324,
|
|
"mean_token_accuracy": 0.192037433385849,
|
|
"num_tokens": 9443807.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 6.267121744155884,
|
|
"epoch": 4.79661859771258,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032921346237426536,
|
|
"loss": 5.6292,
|
|
"mean_token_accuracy": 0.176527039706707,
|
|
"num_tokens": 9452608.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 6.177492380142212,
|
|
"epoch": 4.8015912481352565,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032883435726801635,
|
|
"loss": 5.6397,
|
|
"mean_token_accuracy": 0.179460546374321,
|
|
"num_tokens": 9462233.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 6.230467653274536,
|
|
"epoch": 4.806563898557932,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003284550899800036,
|
|
"loss": 5.6549,
|
|
"mean_token_accuracy": 0.17920474708080292,
|
|
"num_tokens": 9472926.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 6.2170679569244385,
|
|
"epoch": 4.811536548980607,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032807566165281055,
|
|
"loss": 5.5641,
|
|
"mean_token_accuracy": 0.18994470089673995,
|
|
"num_tokens": 9481711.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 6.237385320663452,
|
|
"epoch": 4.816509199403282,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003276960734295054,
|
|
"loss": 5.5324,
|
|
"mean_token_accuracy": 0.18121843039989471,
|
|
"num_tokens": 9492655.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 6.212373399734497,
|
|
"epoch": 4.8214818498259575,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032731632645363856,
|
|
"loss": 5.524,
|
|
"mean_token_accuracy": 0.19520680457353592,
|
|
"num_tokens": 9502852.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 6.152708196640015,
|
|
"epoch": 4.826454500248633,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003269364218692384,
|
|
"loss": 5.4875,
|
|
"mean_token_accuracy": 0.19489496052265168,
|
|
"num_tokens": 9511781.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 6.1883820533752445,
|
|
"epoch": 4.831427150671308,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003265563608208081,
|
|
"loss": 5.5765,
|
|
"mean_token_accuracy": 0.19063158482313156,
|
|
"num_tokens": 9521436.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 6.205262041091919,
|
|
"epoch": 4.836399801093983,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003261761444533221,
|
|
"loss": 5.5196,
|
|
"mean_token_accuracy": 0.19574809223413467,
|
|
"num_tokens": 9531432.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 6.228746747970581,
|
|
"epoch": 4.8413724515166585,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003257957739122231,
|
|
"loss": 5.6801,
|
|
"mean_token_accuracy": 0.18745245635509492,
|
|
"num_tokens": 9540976.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 6.175098609924317,
|
|
"epoch": 4.846345101939334,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032541525034341805,
|
|
"loss": 5.4586,
|
|
"mean_token_accuracy": 0.19249564409255981,
|
|
"num_tokens": 9549566.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 6.128291511535645,
|
|
"epoch": 4.851317752362009,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003250345748932749,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.18461671322584153,
|
|
"num_tokens": 9558937.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 6.132266807556152,
|
|
"epoch": 4.856290402784684,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032465374870861913,
|
|
"loss": 5.5612,
|
|
"mean_token_accuracy": 0.19395255148410798,
|
|
"num_tokens": 9568374.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 6.073661279678345,
|
|
"epoch": 4.8612630532073595,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032427277293673053,
|
|
"loss": 5.4139,
|
|
"mean_token_accuracy": 0.20356135964393615,
|
|
"num_tokens": 9577835.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 6.195731496810913,
|
|
"epoch": 4.866235703630035,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003238916487253393,
|
|
"loss": 5.5904,
|
|
"mean_token_accuracy": 0.18322182148694993,
|
|
"num_tokens": 9586787.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 6.19819712638855,
|
|
"epoch": 4.87120835405271,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003235103772226229,
|
|
"loss": 5.616,
|
|
"mean_token_accuracy": 0.18617940992116927,
|
|
"num_tokens": 9597669.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 6.1902174949646,
|
|
"epoch": 4.876181004475385,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003231289595772026,
|
|
"loss": 5.5042,
|
|
"mean_token_accuracy": 0.19446465075016023,
|
|
"num_tokens": 9607535.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 6.213953638076783,
|
|
"epoch": 4.8811536548980605,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003227473969381398,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.18465341180562972,
|
|
"num_tokens": 9616745.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 6.182901048660279,
|
|
"epoch": 4.886126305320736,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032236569045493283,
|
|
"loss": 5.6567,
|
|
"mean_token_accuracy": 0.17805507630109788,
|
|
"num_tokens": 9626445.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 6.189204549789428,
|
|
"epoch": 4.891098955743411,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032198384127751344,
|
|
"loss": 5.4982,
|
|
"mean_token_accuracy": 0.19778303653001786,
|
|
"num_tokens": 9635083.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 6.219792079925537,
|
|
"epoch": 4.896071606166086,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000321601850556243,
|
|
"loss": 5.675,
|
|
"mean_token_accuracy": 0.1786137819290161,
|
|
"num_tokens": 9645410.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 6.206762218475342,
|
|
"epoch": 4.9010442565887615,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003212197194419095,
|
|
"loss": 5.5968,
|
|
"mean_token_accuracy": 0.1811181202530861,
|
|
"num_tokens": 9654826.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 6.232389974594116,
|
|
"epoch": 4.906016907011437,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003208374490857238,
|
|
"loss": 5.6222,
|
|
"mean_token_accuracy": 0.18120293766260148,
|
|
"num_tokens": 9664609.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 6.0710937023162845,
|
|
"epoch": 4.910989557434112,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003204550406393164,
|
|
"loss": 5.428,
|
|
"mean_token_accuracy": 0.20508671104907988,
|
|
"num_tokens": 9672963.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 6.215498208999634,
|
|
"epoch": 4.915962207856787,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003200724952547335,
|
|
"loss": 5.4858,
|
|
"mean_token_accuracy": 0.19593576043844224,
|
|
"num_tokens": 9681662.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 6.114095401763916,
|
|
"epoch": 4.9209348582794625,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003196898140844341,
|
|
"loss": 5.4273,
|
|
"mean_token_accuracy": 0.20229132920503617,
|
|
"num_tokens": 9693204.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 6.167608022689819,
|
|
"epoch": 4.925907508702139,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031930699828128614,
|
|
"loss": 5.5203,
|
|
"mean_token_accuracy": 0.1903752401471138,
|
|
"num_tokens": 9702964.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 6.2110649108886715,
|
|
"epoch": 4.930880159124813,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003189240489985633,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.19864056408405303,
|
|
"num_tokens": 9712407.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 6.2446229457855225,
|
|
"epoch": 4.935852809547489,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003185409673899412,
|
|
"loss": 5.6051,
|
|
"mean_token_accuracy": 0.18386708647012712,
|
|
"num_tokens": 9722123.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 6.240275621414185,
|
|
"epoch": 4.940825459970164,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00031815775460949426,
|
|
"loss": 5.5504,
|
|
"mean_token_accuracy": 0.1893150895833969,
|
|
"num_tokens": 9732641.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 6.199017906188965,
|
|
"epoch": 4.94579811039284,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003177744118116919,
|
|
"loss": 5.6145,
|
|
"mean_token_accuracy": 0.18610313385725022,
|
|
"num_tokens": 9741498.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 6.231387996673584,
|
|
"epoch": 4.950770760815515,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003173909401513954,
|
|
"loss": 5.6166,
|
|
"mean_token_accuracy": 0.18297646045684815,
|
|
"num_tokens": 9750811.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 6.045255851745606,
|
|
"epoch": 4.95574341123819,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031700734078385416,
|
|
"loss": 5.4368,
|
|
"mean_token_accuracy": 0.20900335907936096,
|
|
"num_tokens": 9761233.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 6.157975339889527,
|
|
"epoch": 4.960716061660865,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031662361486470243,
|
|
"loss": 5.4938,
|
|
"mean_token_accuracy": 0.1877997487783432,
|
|
"num_tokens": 9771258.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 6.10872368812561,
|
|
"epoch": 4.965688712083541,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003162397635499555,
|
|
"loss": 5.5124,
|
|
"mean_token_accuracy": 0.1940624862909317,
|
|
"num_tokens": 9781501.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 6.213992977142334,
|
|
"epoch": 4.970661362506216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031585578799600653,
|
|
"loss": 5.7707,
|
|
"mean_token_accuracy": 0.1749344974756241,
|
|
"num_tokens": 9792488.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.970661362506216,
|
|
"eval_entropy": 6.0814109980236,
|
|
"eval_loss": 6.442910671234131,
|
|
"eval_mean_token_accuracy": 0.1485907713955586,
|
|
"eval_num_tokens": 9792488.0,
|
|
"eval_runtime": 2.1208,
|
|
"eval_samples_per_second": 1633.787,
|
|
"eval_steps_per_second": 204.636,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 6.219845390319824,
|
|
"epoch": 4.975634012928891,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031547168935962304,
|
|
"loss": 5.5363,
|
|
"mean_token_accuracy": 0.18949103355407715,
|
|
"num_tokens": 9802711.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 6.237148237228394,
|
|
"epoch": 4.980606663351566,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003150874687979433,
|
|
"loss": 5.5715,
|
|
"mean_token_accuracy": 0.19228897392749786,
|
|
"num_tokens": 9812668.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 6.150430583953858,
|
|
"epoch": 4.985579313774242,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031470312746847284,
|
|
"loss": 5.5792,
|
|
"mean_token_accuracy": 0.19320937842130662,
|
|
"num_tokens": 9822466.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 6.232890844345093,
|
|
"epoch": 4.990551964196917,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00031431866652908107,
|
|
"loss": 5.5758,
|
|
"mean_token_accuracy": 0.19253599792718887,
|
|
"num_tokens": 9831529.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 6.203870677947998,
|
|
"epoch": 4.995524614619592,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003139340871379977,
|
|
"loss": 5.6342,
|
|
"mean_token_accuracy": 0.1959534093737602,
|
|
"num_tokens": 9842321.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 6.233012570275201,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.0003135493904538093,
|
|
"loss": 5.6412,
|
|
"mean_token_accuracy": 0.18758039673169455,
|
|
"num_tokens": 9850520.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 6.235342788696289,
|
|
"epoch": 5.004972650422675,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003131645776354559,
|
|
"loss": 5.2331,
|
|
"mean_token_accuracy": 0.22322876006364822,
|
|
"num_tokens": 9859880.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 6.177197504043579,
|
|
"epoch": 5.0099453008453505,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003127796498422272,
|
|
"loss": 5.3709,
|
|
"mean_token_accuracy": 0.20105646699666976,
|
|
"num_tokens": 9869409.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 6.147691965103149,
|
|
"epoch": 5.014917951268026,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00031239460823375933,
|
|
"loss": 5.3517,
|
|
"mean_token_accuracy": 0.19571183174848555,
|
|
"num_tokens": 9879558.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 6.233658361434936,
|
|
"epoch": 5.019890601690701,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031200945397003157,
|
|
"loss": 5.3279,
|
|
"mean_token_accuracy": 0.20710447132587434,
|
|
"num_tokens": 9890612.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 6.215064859390258,
|
|
"epoch": 5.024863252113376,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003116241882113623,
|
|
"loss": 5.3414,
|
|
"mean_token_accuracy": 0.20268536806106568,
|
|
"num_tokens": 9900928.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 6.214122343063354,
|
|
"epoch": 5.0298359025360515,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031123881211840565,
|
|
"loss": 5.384,
|
|
"mean_token_accuracy": 0.20253535360097885,
|
|
"num_tokens": 9910627.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 6.217760133743286,
|
|
"epoch": 5.034808552958727,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003108533268521485,
|
|
"loss": 5.3388,
|
|
"mean_token_accuracy": 0.1984658196568489,
|
|
"num_tokens": 9921250.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 6.111049747467041,
|
|
"epoch": 5.039781203381402,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003104677335739065,
|
|
"loss": 5.2543,
|
|
"mean_token_accuracy": 0.2177567884325981,
|
|
"num_tokens": 9931213.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 6.153992414474487,
|
|
"epoch": 5.044753853804077,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00031008203344532067,
|
|
"loss": 5.3325,
|
|
"mean_token_accuracy": 0.2065000206232071,
|
|
"num_tokens": 9941385.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 6.1534301280975345,
|
|
"epoch": 5.0497265042267525,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000309696227628354,
|
|
"loss": 5.2136,
|
|
"mean_token_accuracy": 0.21375870555639268,
|
|
"num_tokens": 9951547.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 6.192005443572998,
|
|
"epoch": 5.054699154649428,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030931031728528754,
|
|
"loss": 5.3776,
|
|
"mean_token_accuracy": 0.19805853217840194,
|
|
"num_tokens": 9960855.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 6.201622152328492,
|
|
"epoch": 5.059671805072103,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003089243035787178,
|
|
"loss": 5.4212,
|
|
"mean_token_accuracy": 0.19672441333532334,
|
|
"num_tokens": 9970116.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.968948411941528,
|
|
"epoch": 5.064644455494779,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00030853818767155225,
|
|
"loss": 5.142,
|
|
"mean_token_accuracy": 0.22970665395259857,
|
|
"num_tokens": 9980488.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 6.172068691253662,
|
|
"epoch": 5.069617105917454,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003081519707270065,
|
|
"loss": 5.4042,
|
|
"mean_token_accuracy": 0.1991195023059845,
|
|
"num_tokens": 9990677.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 6.0170081615447994,
|
|
"epoch": 5.07458975634013,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030776565390860017,
|
|
"loss": 5.0984,
|
|
"mean_token_accuracy": 0.2232486978173256,
|
|
"num_tokens": 10000218.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 6.161630010604858,
|
|
"epoch": 5.079562406762805,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003073792383801543,
|
|
"loss": 5.3004,
|
|
"mean_token_accuracy": 0.2147019848227501,
|
|
"num_tokens": 10008720.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 6.166669845581055,
|
|
"epoch": 5.08453505718548,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030699272530578697,
|
|
"loss": 5.3675,
|
|
"mean_token_accuracy": 0.21088258624076844,
|
|
"num_tokens": 10018648.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 6.129061460494995,
|
|
"epoch": 5.089507707608155,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030660611584991013,
|
|
"loss": 5.2862,
|
|
"mean_token_accuracy": 0.2097887396812439,
|
|
"num_tokens": 10028639.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 6.155539846420288,
|
|
"epoch": 5.094480358030831,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030621941117722615,
|
|
"loss": 5.3353,
|
|
"mean_token_accuracy": 0.20747511237859725,
|
|
"num_tokens": 10038136.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 6.103600645065308,
|
|
"epoch": 5.099453008453506,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003058326124527243,
|
|
"loss": 5.3103,
|
|
"mean_token_accuracy": 0.20677597671747208,
|
|
"num_tokens": 10048503.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 6.114904356002808,
|
|
"epoch": 5.104425658876181,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003054457208416771,
|
|
"loss": 5.2737,
|
|
"mean_token_accuracy": 0.213410884141922,
|
|
"num_tokens": 10059088.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 6.135048484802246,
|
|
"epoch": 5.109398309298856,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00030505873750963705,
|
|
"loss": 5.2888,
|
|
"mean_token_accuracy": 0.20730531215667725,
|
|
"num_tokens": 10069655.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 6.197358894348144,
|
|
"epoch": 5.114370959721532,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003046716636224327,
|
|
"loss": 5.2969,
|
|
"mean_token_accuracy": 0.20810099989175795,
|
|
"num_tokens": 10079748.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 6.166021919250488,
|
|
"epoch": 5.119343610144207,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003042845003461657,
|
|
"loss": 5.2966,
|
|
"mean_token_accuracy": 0.2039845570921898,
|
|
"num_tokens": 10088877.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 6.201730251312256,
|
|
"epoch": 5.124316260566882,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030389724884720664,
|
|
"loss": 5.389,
|
|
"mean_token_accuracy": 0.19809055477380752,
|
|
"num_tokens": 10098699.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.869723510742188,
|
|
"epoch": 5.129288910989557,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00030350991029219236,
|
|
"loss": 5.0026,
|
|
"mean_token_accuracy": 0.2490513503551483,
|
|
"num_tokens": 10110084.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 6.0983250617980955,
|
|
"epoch": 5.134261561412233,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030312248584802166,
|
|
"loss": 5.3618,
|
|
"mean_token_accuracy": 0.20196272432804108,
|
|
"num_tokens": 10119553.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 6.206573963165283,
|
|
"epoch": 5.139234211834908,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003027349766818521,
|
|
"loss": 5.2981,
|
|
"mean_token_accuracy": 0.20535645186901091,
|
|
"num_tokens": 10128332.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 6.142053079605103,
|
|
"epoch": 5.144206862257583,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003023473839610965,
|
|
"loss": 5.3859,
|
|
"mean_token_accuracy": 0.19929342865943908,
|
|
"num_tokens": 10138109.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 6.205005645751953,
|
|
"epoch": 5.149179512680258,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003019597088534195,
|
|
"loss": 5.4637,
|
|
"mean_token_accuracy": 0.19362542331218718,
|
|
"num_tokens": 10148103.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 6.286076402664184,
|
|
"epoch": 5.154152163102934,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003015719525267338,
|
|
"loss": 5.4653,
|
|
"mean_token_accuracy": 0.19011985510587692,
|
|
"num_tokens": 10158189.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 6.192364358901978,
|
|
"epoch": 5.159124813525609,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003011841161491969,
|
|
"loss": 5.3908,
|
|
"mean_token_accuracy": 0.20639271438121795,
|
|
"num_tokens": 10169048.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 6.076186609268189,
|
|
"epoch": 5.164097463948284,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030079620088920727,
|
|
"loss": 5.3148,
|
|
"mean_token_accuracy": 0.21019342243671418,
|
|
"num_tokens": 10178067.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 6.155745792388916,
|
|
"epoch": 5.169070114370959,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003004082079154013,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.1999416545033455,
|
|
"num_tokens": 10186952.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 6.190444135665894,
|
|
"epoch": 5.174042764793635,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030002013839664925,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.19716678708791732,
|
|
"num_tokens": 10196699.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 6.149278163909912,
|
|
"epoch": 5.17901541521631,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002996319935020522,
|
|
"loss": 5.2677,
|
|
"mean_token_accuracy": 0.20509596765041352,
|
|
"num_tokens": 10206198.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 6.152007007598877,
|
|
"epoch": 5.183988065638985,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002992437744009381,
|
|
"loss": 5.3579,
|
|
"mean_token_accuracy": 0.19933388382196426,
|
|
"num_tokens": 10215698.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 6.242123365402222,
|
|
"epoch": 5.188960716061661,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00029885548226285855,
|
|
"loss": 5.4218,
|
|
"mean_token_accuracy": 0.19652938842773438,
|
|
"num_tokens": 10225732.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 6.17102165222168,
|
|
"epoch": 5.193933366484337,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002984671182575851,
|
|
"loss": 5.3552,
|
|
"mean_token_accuracy": 0.20912957936525345,
|
|
"num_tokens": 10235691.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 6.1554583549499515,
|
|
"epoch": 5.198906016907012,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002980786835551061,
|
|
"loss": 5.3684,
|
|
"mean_token_accuracy": 0.2126176744699478,
|
|
"num_tokens": 10246066.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 6.203218936920166,
|
|
"epoch": 5.203878667329687,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002976901793256225,
|
|
"loss": 5.3059,
|
|
"mean_token_accuracy": 0.20557932406663895,
|
|
"num_tokens": 10254968.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 6.162584352493286,
|
|
"epoch": 5.208851317752362,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000297301606739545,
|
|
"loss": 5.3349,
|
|
"mean_token_accuracy": 0.19974741488695144,
|
|
"num_tokens": 10263800.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 6.14858717918396,
|
|
"epoch": 5.213823968175038,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00029691296696748994,
|
|
"loss": 5.3975,
|
|
"mean_token_accuracy": 0.19627835750579833,
|
|
"num_tokens": 10274044.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 6.179690551757813,
|
|
"epoch": 5.218796618597713,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00029652426118027635,
|
|
"loss": 5.4175,
|
|
"mean_token_accuracy": 0.19749944508075715,
|
|
"num_tokens": 10284429.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 6.1492376804351805,
|
|
"epoch": 5.223769269020388,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029613549054892206,
|
|
"loss": 5.4048,
|
|
"mean_token_accuracy": 0.1995590388774872,
|
|
"num_tokens": 10293968.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.974618816375733,
|
|
"epoch": 5.228741919443063,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002957466562446401,
|
|
"loss": 5.1579,
|
|
"mean_token_accuracy": 0.22444677501916885,
|
|
"num_tokens": 10303497.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 6.170616626739502,
|
|
"epoch": 5.233714569865739,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002953577594388355,
|
|
"loss": 5.4105,
|
|
"mean_token_accuracy": 0.19241058230400085,
|
|
"num_tokens": 10312766.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 6.140893983840942,
|
|
"epoch": 5.238687220288414,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00029496880130310145,
|
|
"loss": 5.2868,
|
|
"mean_token_accuracy": 0.20914850383996964,
|
|
"num_tokens": 10323604.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 6.155068445205688,
|
|
"epoch": 5.243659870711089,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002945797830092161,
|
|
"loss": 5.3511,
|
|
"mean_token_accuracy": 0.20059113204479218,
|
|
"num_tokens": 10333620.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 6.116631889343262,
|
|
"epoch": 5.248632521133764,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002941907057291386,
|
|
"loss": 5.3629,
|
|
"mean_token_accuracy": 0.20161808133125306,
|
|
"num_tokens": 10344078.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 6.229410028457641,
|
|
"epoch": 5.25360517155644,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00029380157063500594,
|
|
"loss": 5.3173,
|
|
"mean_token_accuracy": 0.20401767790317535,
|
|
"num_tokens": 10353254.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 6.105338001251221,
|
|
"epoch": 5.258577821979115,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002934123788991291,
|
|
"loss": 5.3437,
|
|
"mean_token_accuracy": 0.20693225413560867,
|
|
"num_tokens": 10362310.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 6.183104228973389,
|
|
"epoch": 5.26355047240179,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029302313169399,
|
|
"loss": 5.5488,
|
|
"mean_token_accuracy": 0.18971027433872223,
|
|
"num_tokens": 10373252.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 6.161838006973267,
|
|
"epoch": 5.268523122824465,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002926338301922376,
|
|
"loss": 5.3453,
|
|
"mean_token_accuracy": 0.20853369385004045,
|
|
"num_tokens": 10381682.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 6.241167449951172,
|
|
"epoch": 5.273495773247141,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029224447556668417,
|
|
"loss": 5.5138,
|
|
"mean_token_accuracy": 0.18879543244838715,
|
|
"num_tokens": 10391567.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 6.204581165313721,
|
|
"epoch": 5.278468423669816,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002918550689903022,
|
|
"loss": 5.4353,
|
|
"mean_token_accuracy": 0.1943253308534622,
|
|
"num_tokens": 10401820.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 6.184020280838013,
|
|
"epoch": 5.283441074092491,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00029146561163622076,
|
|
"loss": 5.3882,
|
|
"mean_token_accuracy": 0.20445790588855745,
|
|
"num_tokens": 10413378.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 6.115431499481201,
|
|
"epoch": 5.288413724515166,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002910761046777218,
|
|
"loss": 5.306,
|
|
"mean_token_accuracy": 0.21136901676654815,
|
|
"num_tokens": 10423447.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 6.1304261684417725,
|
|
"epoch": 5.293386374937842,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00029068654928823666,
|
|
"loss": 5.2627,
|
|
"mean_token_accuracy": 0.20930971652269365,
|
|
"num_tokens": 10433785.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 6.124007654190064,
|
|
"epoch": 5.298359025360517,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00029029694664134284,
|
|
"loss": 5.3883,
|
|
"mean_token_accuracy": 0.19006958305835725,
|
|
"num_tokens": 10444748.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 6.221176052093506,
|
|
"epoch": 5.303331675783192,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002899072979107597,
|
|
"loss": 5.45,
|
|
"mean_token_accuracy": 0.18771999925374985,
|
|
"num_tokens": 10455778.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 6.172452211380005,
|
|
"epoch": 5.308304326205867,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002895176042703461,
|
|
"loss": 5.4242,
|
|
"mean_token_accuracy": 0.19776239842176438,
|
|
"num_tokens": 10465915.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 6.124274826049804,
|
|
"epoch": 5.3132769766285435,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00028912786689409554,
|
|
"loss": 5.2226,
|
|
"mean_token_accuracy": 0.22274541556835176,
|
|
"num_tokens": 10475838.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.948004722595215,
|
|
"epoch": 5.318249627051218,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002887380869561338,
|
|
"loss": 5.1262,
|
|
"mean_token_accuracy": 0.2259490370750427,
|
|
"num_tokens": 10485719.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 6.169914245605469,
|
|
"epoch": 5.323222277473894,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002883482656307145,
|
|
"loss": 5.3265,
|
|
"mean_token_accuracy": 0.20450772643089293,
|
|
"num_tokens": 10494695.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 6.221499347686768,
|
|
"epoch": 5.328194927896569,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002879584040922162,
|
|
"loss": 5.4436,
|
|
"mean_token_accuracy": 0.19904973357915878,
|
|
"num_tokens": 10505356.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 6.12182583808899,
|
|
"epoch": 5.3331675783192445,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002875685035151386,
|
|
"loss": 5.2282,
|
|
"mean_token_accuracy": 0.20691704452037812,
|
|
"num_tokens": 10515118.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 6.044535970687866,
|
|
"epoch": 5.33814022874192,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028717856507409886,
|
|
"loss": 5.3438,
|
|
"mean_token_accuracy": 0.20523097068071366,
|
|
"num_tokens": 10525713.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 6.1392632007598875,
|
|
"epoch": 5.343112879164595,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002867885899438281,
|
|
"loss": 5.2912,
|
|
"mean_token_accuracy": 0.21036715507507325,
|
|
"num_tokens": 10534363.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 6.062277030944824,
|
|
"epoch": 5.34808552958727,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002863985792991683,
|
|
"loss": 5.1912,
|
|
"mean_token_accuracy": 0.2142636686563492,
|
|
"num_tokens": 10544302.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 6.130734252929687,
|
|
"epoch": 5.3530581800099455,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002860085343150683,
|
|
"loss": 5.363,
|
|
"mean_token_accuracy": 0.20206487774848939,
|
|
"num_tokens": 10553742.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 6.186791849136353,
|
|
"epoch": 5.358030830432621,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00028561845616658033,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.18907036632299423,
|
|
"num_tokens": 10564061.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 6.17963604927063,
|
|
"epoch": 5.363003480855296,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00028522834602885656,
|
|
"loss": 5.4331,
|
|
"mean_token_accuracy": 0.2011225402355194,
|
|
"num_tokens": 10573439.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 6.221355056762695,
|
|
"epoch": 5.367976131277971,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002848382050771455,
|
|
"loss": 5.4009,
|
|
"mean_token_accuracy": 0.19981967955827712,
|
|
"num_tokens": 10583694.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 6.155561304092407,
|
|
"epoch": 5.3729487817006465,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002844480344867886,
|
|
"loss": 5.452,
|
|
"mean_token_accuracy": 0.19610781222581863,
|
|
"num_tokens": 10593432.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.933334445953369,
|
|
"epoch": 5.377921432123322,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002840578354332165,
|
|
"loss": 5.1174,
|
|
"mean_token_accuracy": 0.23127251565456391,
|
|
"num_tokens": 10603050.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 6.12565860748291,
|
|
"epoch": 5.382894082545997,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002836676090919456,
|
|
"loss": 5.3697,
|
|
"mean_token_accuracy": 0.204098679125309,
|
|
"num_tokens": 10613842.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 6.155111360549927,
|
|
"epoch": 5.387866732968672,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002832773566385746,
|
|
"loss": 5.3615,
|
|
"mean_token_accuracy": 0.20062829107046126,
|
|
"num_tokens": 10624220.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 6.108516788482666,
|
|
"epoch": 5.3928393833913475,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00028288707924878056,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.21752716600894928,
|
|
"num_tokens": 10633797.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 6.085815000534057,
|
|
"epoch": 5.397812033814023,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002824967780983162,
|
|
"loss": 5.2921,
|
|
"mean_token_accuracy": 0.20842369943857192,
|
|
"num_tokens": 10642718.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 6.2328814506530765,
|
|
"epoch": 5.402784684236698,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028210645436300524,
|
|
"loss": 5.4609,
|
|
"mean_token_accuracy": 0.18718303591012955,
|
|
"num_tokens": 10653733.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 6.1204798221588135,
|
|
"epoch": 5.407757334659373,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028171610921873995,
|
|
"loss": 5.3627,
|
|
"mean_token_accuracy": 0.19833193421363832,
|
|
"num_tokens": 10663238.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 6.145351600646973,
|
|
"epoch": 5.4127299850820485,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002813257438414766,
|
|
"loss": 5.414,
|
|
"mean_token_accuracy": 0.19865717589855195,
|
|
"num_tokens": 10674037.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 6.178290319442749,
|
|
"epoch": 5.417702635504724,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000280935359407233,
|
|
"loss": 5.4123,
|
|
"mean_token_accuracy": 0.19508548825979233,
|
|
"num_tokens": 10684394.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 6.144301700592041,
|
|
"epoch": 5.422675285927399,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002805449570920838,
|
|
"loss": 5.4377,
|
|
"mean_token_accuracy": 0.1991276904940605,
|
|
"num_tokens": 10693396.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 6.116199111938476,
|
|
"epoch": 5.427647936350074,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028015453807215786,
|
|
"loss": 5.3718,
|
|
"mean_token_accuracy": 0.20383760929107667,
|
|
"num_tokens": 10703643.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 6.143051528930664,
|
|
"epoch": 5.4326205867727495,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00027976410352363416,
|
|
"loss": 5.2971,
|
|
"mean_token_accuracy": 0.20340121686458587,
|
|
"num_tokens": 10712883.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 6.114397716522217,
|
|
"epoch": 5.437593237195425,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00027937365462273864,
|
|
"loss": 5.2847,
|
|
"mean_token_accuracy": 0.20064962059259414,
|
|
"num_tokens": 10721926.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 6.115090036392212,
|
|
"epoch": 5.4425658876181,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00027898319254574043,
|
|
"loss": 5.3538,
|
|
"mean_token_accuracy": 0.201101590692997,
|
|
"num_tokens": 10731772.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 6.190639495849609,
|
|
"epoch": 5.447538538040776,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002785927184689483,
|
|
"loss": 5.4999,
|
|
"mean_token_accuracy": 0.19772129952907563,
|
|
"num_tokens": 10742815.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 6.153209447860718,
|
|
"epoch": 5.452511188463451,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002782022335687073,
|
|
"loss": 5.3242,
|
|
"mean_token_accuracy": 0.2126567006111145,
|
|
"num_tokens": 10753958.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 6.202142810821533,
|
|
"epoch": 5.457483838886127,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00027781173902139485,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.19318765699863433,
|
|
"num_tokens": 10763808.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 6.152526903152466,
|
|
"epoch": 5.462456489308802,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002774212360034176,
|
|
"loss": 5.3978,
|
|
"mean_token_accuracy": 0.2018960729241371,
|
|
"num_tokens": 10773688.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 6.176838684082031,
|
|
"epoch": 5.467429139731477,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002770307256912078,
|
|
"loss": 5.4117,
|
|
"mean_token_accuracy": 0.20072914510965348,
|
|
"num_tokens": 10783902.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.975389528274536,
|
|
"epoch": 5.472401790154152,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027664020926121944,
|
|
"loss": 5.0904,
|
|
"mean_token_accuracy": 0.22805523872375488,
|
|
"num_tokens": 10794138.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 6.118174886703491,
|
|
"epoch": 5.477374440576828,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002762496878899251,
|
|
"loss": 5.3845,
|
|
"mean_token_accuracy": 0.20123378336429595,
|
|
"num_tokens": 10804599.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 6.158814334869385,
|
|
"epoch": 5.482347090999503,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027585916275381226,
|
|
"loss": 5.3505,
|
|
"mean_token_accuracy": 0.20945720225572587,
|
|
"num_tokens": 10813632.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 6.152644872665405,
|
|
"epoch": 5.487319741422178,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002754686350293797,
|
|
"loss": 5.4356,
|
|
"mean_token_accuracy": 0.19604238420724868,
|
|
"num_tokens": 10824287.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 6.171857452392578,
|
|
"epoch": 5.492292391844853,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002750781058931339,
|
|
"loss": 5.4508,
|
|
"mean_token_accuracy": 0.20391667783260345,
|
|
"num_tokens": 10834079.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 6.163483953475952,
|
|
"epoch": 5.497265042267529,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002746875765215857,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.19636480659246444,
|
|
"num_tokens": 10843029.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 6.136315107345581,
|
|
"epoch": 5.502237692690204,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027429704809124665,
|
|
"loss": 5.3734,
|
|
"mean_token_accuracy": 0.20156598538160325,
|
|
"num_tokens": 10852749.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 6.046795225143432,
|
|
"epoch": 5.507210343112879,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002739065217786255,
|
|
"loss": 5.2695,
|
|
"mean_token_accuracy": 0.21237550526857377,
|
|
"num_tokens": 10862911.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 6.1784319400787355,
|
|
"epoch": 5.512182993535554,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002735159987602246,
|
|
"loss": 5.318,
|
|
"mean_token_accuracy": 0.20769803375005721,
|
|
"num_tokens": 10872389.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 6.119740152359009,
|
|
"epoch": 5.51715564395823,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002731254802125362,
|
|
"loss": 5.2784,
|
|
"mean_token_accuracy": 0.20771167427301407,
|
|
"num_tokens": 10881280.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 6.040499925613403,
|
|
"epoch": 5.522128294380905,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027273496731203945,
|
|
"loss": 5.4042,
|
|
"mean_token_accuracy": 0.20619335323572158,
|
|
"num_tokens": 10891154.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 6.119177007675171,
|
|
"epoch": 5.52710094480358,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000272344461235196,
|
|
"loss": 5.3026,
|
|
"mean_token_accuracy": 0.21196686178445817,
|
|
"num_tokens": 10900029.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.945767641067505,
|
|
"epoch": 5.532073595226255,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00027195396315844733,
|
|
"loss": 5.1637,
|
|
"mean_token_accuracy": 0.23316535204648972,
|
|
"num_tokens": 10909616.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.986102962493897,
|
|
"epoch": 5.537046245648931,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002715634742582107,
|
|
"loss": 5.1235,
|
|
"mean_token_accuracy": 0.22949036806821824,
|
|
"num_tokens": 10920754.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 6.109731817245484,
|
|
"epoch": 5.542018896071606,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00027117299571087576,
|
|
"loss": 5.3499,
|
|
"mean_token_accuracy": 0.20222352892160417,
|
|
"num_tokens": 10929918.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 6.150139999389649,
|
|
"epoch": 5.546991546494281,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002707825286928008,
|
|
"loss": 5.3684,
|
|
"mean_token_accuracy": 0.20569818019866942,
|
|
"num_tokens": 10938932.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 6.112952041625976,
|
|
"epoch": 5.551964196916956,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002703920743803097,
|
|
"loss": 5.282,
|
|
"mean_token_accuracy": 0.20157541632652282,
|
|
"num_tokens": 10948896.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 6.172693729400635,
|
|
"epoch": 5.556936847339632,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027000163394968763,
|
|
"loss": 5.4096,
|
|
"mean_token_accuracy": 0.1983227014541626,
|
|
"num_tokens": 10958245.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 6.1288656234741214,
|
|
"epoch": 5.561909497762308,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026961120857717843,
|
|
"loss": 5.3688,
|
|
"mean_token_accuracy": 0.20593727976083756,
|
|
"num_tokens": 10967917.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 6.125351285934448,
|
|
"epoch": 5.566882148184982,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00026922079943898016,
|
|
"loss": 5.3541,
|
|
"mean_token_accuracy": 0.20491524636745453,
|
|
"num_tokens": 10977062.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 6.19562406539917,
|
|
"epoch": 5.571854798607658,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002688304077112421,
|
|
"loss": 5.4637,
|
|
"mean_token_accuracy": 0.1982416495680809,
|
|
"num_tokens": 10987009.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 6.123086786270141,
|
|
"epoch": 5.576827449030333,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002684400345700613,
|
|
"loss": 5.2244,
|
|
"mean_token_accuracy": 0.21782220155000687,
|
|
"num_tokens": 10996077.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 6.136792230606079,
|
|
"epoch": 5.581800099453009,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002680496811914786,
|
|
"loss": 5.313,
|
|
"mean_token_accuracy": 0.2025127276778221,
|
|
"num_tokens": 11005240.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 6.118552827835083,
|
|
"epoch": 5.586772749875684,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026765934875147513,
|
|
"loss": 5.4426,
|
|
"mean_token_accuracy": 0.19489303827285767,
|
|
"num_tokens": 11014373.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.969034433364868,
|
|
"epoch": 5.591745400298359,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002672690384259693,
|
|
"loss": 5.1566,
|
|
"mean_token_accuracy": 0.2366982638835907,
|
|
"num_tokens": 11024676.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 6.084133625030518,
|
|
"epoch": 5.596718050721035,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026687875139081287,
|
|
"loss": 5.3356,
|
|
"mean_token_accuracy": 0.20485207736492156,
|
|
"num_tokens": 11033722.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 6.110644769668579,
|
|
"epoch": 5.60169070114371,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026648848882178703,
|
|
"loss": 5.3566,
|
|
"mean_token_accuracy": 0.2080860033631325,
|
|
"num_tokens": 11043405.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 6.10286283493042,
|
|
"epoch": 5.606663351566385,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002660982518945998,
|
|
"loss": 5.355,
|
|
"mean_token_accuracy": 0.20449019521474837,
|
|
"num_tokens": 11053577.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 6.107110929489136,
|
|
"epoch": 5.61163600198906,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026570804178488146,
|
|
"loss": 5.331,
|
|
"mean_token_accuracy": 0.2056924283504486,
|
|
"num_tokens": 11062890.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.9989190101623535,
|
|
"epoch": 5.616608652411736,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002653178596681819,
|
|
"loss": 5.1971,
|
|
"mean_token_accuracy": 0.22585684508085252,
|
|
"num_tokens": 11073445.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 6.091164875030517,
|
|
"epoch": 5.621581302834411,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002649277067199665,
|
|
"loss": 5.3779,
|
|
"mean_token_accuracy": 0.20058325678110123,
|
|
"num_tokens": 11083159.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 6.165211868286133,
|
|
"epoch": 5.626553953257086,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002645375841156127,
|
|
"loss": 5.3806,
|
|
"mean_token_accuracy": 0.19392533749341964,
|
|
"num_tokens": 11093417.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 6.1673236846923825,
|
|
"epoch": 5.631526603679761,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002641474930304067,
|
|
"loss": 5.3487,
|
|
"mean_token_accuracy": 0.2081523656845093,
|
|
"num_tokens": 11102775.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 6.242355108261108,
|
|
"epoch": 5.636499254102437,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002637574346395396,
|
|
"loss": 5.4343,
|
|
"mean_token_accuracy": 0.1959111660718918,
|
|
"num_tokens": 11111952.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 6.095498323440552,
|
|
"epoch": 5.641471904525112,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000263367410118104,
|
|
"loss": 5.2267,
|
|
"mean_token_accuracy": 0.21134291738271713,
|
|
"num_tokens": 11120391.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 6.156837272644043,
|
|
"epoch": 5.646444554947787,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002629774206410906,
|
|
"loss": 5.3932,
|
|
"mean_token_accuracy": 0.1945261299610138,
|
|
"num_tokens": 11129905.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 6.092933607101441,
|
|
"epoch": 5.651417205370462,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002625874673833844,
|
|
"loss": 5.3834,
|
|
"mean_token_accuracy": 0.20624370723962784,
|
|
"num_tokens": 11139068.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 6.131637525558472,
|
|
"epoch": 5.656389855793138,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002621975515197613,
|
|
"loss": 5.297,
|
|
"mean_token_accuracy": 0.21291419267654418,
|
|
"num_tokens": 11147718.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 6.167872714996338,
|
|
"epoch": 5.661362506215813,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026180767422488466,
|
|
"loss": 5.4696,
|
|
"mean_token_accuracy": 0.19092469215393065,
|
|
"num_tokens": 11157621.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 6.204186153411865,
|
|
"epoch": 5.666335156638488,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026141783667330156,
|
|
"loss": 5.3674,
|
|
"mean_token_accuracy": 0.2008470341563225,
|
|
"num_tokens": 11168061.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 6.212676095962524,
|
|
"epoch": 5.671307807061163,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002610280400394393,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.19179804921150206,
|
|
"num_tokens": 11177240.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 6.133373641967774,
|
|
"epoch": 5.676280457483839,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00026063828549760195,
|
|
"loss": 5.4723,
|
|
"mean_token_accuracy": 0.20016269385814667,
|
|
"num_tokens": 11187280.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 6.072475528717041,
|
|
"epoch": 5.681253107906514,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002602485742219667,
|
|
"loss": 5.3106,
|
|
"mean_token_accuracy": 0.20671178996562958,
|
|
"num_tokens": 11197207.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 6.160987663269043,
|
|
"epoch": 5.686225758329189,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025985890738658077,
|
|
"loss": 5.4283,
|
|
"mean_token_accuracy": 0.1919071465730667,
|
|
"num_tokens": 11206373.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 6.147854852676391,
|
|
"epoch": 5.691198408751864,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000259469286165357,
|
|
"loss": 5.374,
|
|
"mean_token_accuracy": 0.1951242595911026,
|
|
"num_tokens": 11216799.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 6.166343402862549,
|
|
"epoch": 5.6961710591745405,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025907971173207107,
|
|
"loss": 5.4223,
|
|
"mean_token_accuracy": 0.1886846676468849,
|
|
"num_tokens": 11226968.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 6.136632490158081,
|
|
"epoch": 5.701143709597215,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002586901852603576,
|
|
"loss": 5.4151,
|
|
"mean_token_accuracy": 0.20149416327476502,
|
|
"num_tokens": 11235741.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 6.151400566101074,
|
|
"epoch": 5.706116360019891,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002583007079237069,
|
|
"loss": 5.3925,
|
|
"mean_token_accuracy": 0.19910222589969634,
|
|
"num_tokens": 11245804.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 6.144797992706299,
|
|
"epoch": 5.711089010442566,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025791128089546107,
|
|
"loss": 5.442,
|
|
"mean_token_accuracy": 0.19577786922454835,
|
|
"num_tokens": 11256304.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 6.171363067626953,
|
|
"epoch": 5.7160616608652415,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025752190534881065,
|
|
"loss": 5.431,
|
|
"mean_token_accuracy": 0.19361691325902938,
|
|
"num_tokens": 11265800.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 6.213949155807495,
|
|
"epoch": 5.721034311287917,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00025713258245679127,
|
|
"loss": 5.422,
|
|
"mean_token_accuracy": 0.19792611300945281,
|
|
"num_tokens": 11275983.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 6.188958930969238,
|
|
"epoch": 5.726006961710592,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00025674331339227966,
|
|
"loss": 5.3704,
|
|
"mean_token_accuracy": 0.20927391648292543,
|
|
"num_tokens": 11286375.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 6.181358671188354,
|
|
"epoch": 5.730979612133267,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00025635409932799074,
|
|
"loss": 5.4308,
|
|
"mean_token_accuracy": 0.2002880394458771,
|
|
"num_tokens": 11297359.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 6.159050750732422,
|
|
"epoch": 5.7359522625559425,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002559649414364735,
|
|
"loss": 5.4718,
|
|
"mean_token_accuracy": 0.19463833272457123,
|
|
"num_tokens": 11307235.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 6.2275408744812015,
|
|
"epoch": 5.740924912978618,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002555758408901076,
|
|
"loss": 5.4338,
|
|
"mean_token_accuracy": 0.1883504033088684,
|
|
"num_tokens": 11316618.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 6.2277122974395756,
|
|
"epoch": 5.745897563401293,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00025518679886110023,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.19299621880054474,
|
|
"num_tokens": 11327178.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.978844261169433,
|
|
"epoch": 5.750870213823968,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002547978165214822,
|
|
"loss": 5.1743,
|
|
"mean_token_accuracy": 0.2238128364086151,
|
|
"num_tokens": 11337788.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 6.1054689407348635,
|
|
"epoch": 5.7558428642466435,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00025440889504310416,
|
|
"loss": 5.31,
|
|
"mean_token_accuracy": 0.21586615294218064,
|
|
"num_tokens": 11346958.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 6.168096399307251,
|
|
"epoch": 5.760815514669319,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00025402003559763393,
|
|
"loss": 5.3605,
|
|
"mean_token_accuracy": 0.20293146520853042,
|
|
"num_tokens": 11355817.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 6.140889692306518,
|
|
"epoch": 5.765788165091994,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00025363123935655214,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.2027192696928978,
|
|
"num_tokens": 11366557.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 6.104872560501098,
|
|
"epoch": 5.770760815514669,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002532425074911491,
|
|
"loss": 5.431,
|
|
"mean_token_accuracy": 0.19543099403381348,
|
|
"num_tokens": 11376931.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 6.173142623901367,
|
|
"epoch": 5.7757334659373445,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002528538411725212,
|
|
"loss": 5.4518,
|
|
"mean_token_accuracy": 0.19278985261917114,
|
|
"num_tokens": 11387430.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 6.1260950565338135,
|
|
"epoch": 5.78070611636002,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002524652415715672,
|
|
"loss": 5.3057,
|
|
"mean_token_accuracy": 0.20717668980360032,
|
|
"num_tokens": 11396957.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 6.182759618759155,
|
|
"epoch": 5.785678766782695,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002520767098589852,
|
|
"loss": 5.3213,
|
|
"mean_token_accuracy": 0.20218172967433928,
|
|
"num_tokens": 11405313.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 6.171600341796875,
|
|
"epoch": 5.79065141720537,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002516882472052683,
|
|
"loss": 5.442,
|
|
"mean_token_accuracy": 0.19869366884231568,
|
|
"num_tokens": 11414437.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 6.190905570983887,
|
|
"epoch": 5.7956240676280455,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002512998547807022,
|
|
"loss": 5.5315,
|
|
"mean_token_accuracy": 0.19301831722259521,
|
|
"num_tokens": 11425498.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 6.2501952171325685,
|
|
"epoch": 5.800596718050721,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00025091153375536026,
|
|
"loss": 5.4597,
|
|
"mean_token_accuracy": 0.1908356264233589,
|
|
"num_tokens": 11435333.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.988981628417969,
|
|
"epoch": 5.805569368473396,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025052328529910135,
|
|
"loss": 5.175,
|
|
"mean_token_accuracy": 0.22359997928142547,
|
|
"num_tokens": 11445366.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 6.0756199836730955,
|
|
"epoch": 5.810542018896071,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002501351105815655,
|
|
"loss": 5.3012,
|
|
"mean_token_accuracy": 0.21215294003486634,
|
|
"num_tokens": 11455229.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 6.1649621486663815,
|
|
"epoch": 5.8155146693187465,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024974701077217057,
|
|
"loss": 5.3899,
|
|
"mean_token_accuracy": 0.20119802355766297,
|
|
"num_tokens": 11464789.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 6.144442939758301,
|
|
"epoch": 5.820487319741423,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00024935898704010875,
|
|
"loss": 5.3677,
|
|
"mean_token_accuracy": 0.20447389036417007,
|
|
"num_tokens": 11473694.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 6.182764482498169,
|
|
"epoch": 5.825459970164097,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00024897104055434304,
|
|
"loss": 5.4282,
|
|
"mean_token_accuracy": 0.19832981675863265,
|
|
"num_tokens": 11482701.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 6.1475221633911135,
|
|
"epoch": 5.830432620586773,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024858317248360387,
|
|
"loss": 5.4569,
|
|
"mean_token_accuracy": 0.19704259634017945,
|
|
"num_tokens": 11493917.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 6.1360126495361325,
|
|
"epoch": 5.835405271009448,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002481953839963851,
|
|
"loss": 5.2823,
|
|
"mean_token_accuracy": 0.2085644409060478,
|
|
"num_tokens": 11504001.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 6.175186634063721,
|
|
"epoch": 5.840377921432124,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002478076762609413,
|
|
"loss": 5.4289,
|
|
"mean_token_accuracy": 0.20027776062488556,
|
|
"num_tokens": 11513010.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 6.178004264831543,
|
|
"epoch": 5.845350571854799,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002474200504452834,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.20625869929790497,
|
|
"num_tokens": 11522190.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 6.175077342987061,
|
|
"epoch": 5.850323222277474,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024703250771717576,
|
|
"loss": 5.4228,
|
|
"mean_token_accuracy": 0.2007787525653839,
|
|
"num_tokens": 11531180.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 6.199365091323853,
|
|
"epoch": 5.855295872700149,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024664504924413224,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.19782548546791076,
|
|
"num_tokens": 11540829.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 6.195809173583984,
|
|
"epoch": 5.860268523122825,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002462576761934129,
|
|
"loss": 5.3671,
|
|
"mean_token_accuracy": 0.19192909449338913,
|
|
"num_tokens": 11549118.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 6.156947803497315,
|
|
"epoch": 5.8652411735455,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002458703897320208,
|
|
"loss": 5.3698,
|
|
"mean_token_accuracy": 0.20446102023124696,
|
|
"num_tokens": 11557942.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 6.116778135299683,
|
|
"epoch": 5.870213823968175,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024548319102669745,
|
|
"loss": 5.3335,
|
|
"mean_token_accuracy": 0.20300647914409636,
|
|
"num_tokens": 11567273.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 6.1357728958129885,
|
|
"epoch": 5.87518647439085,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024509608124392065,
|
|
"loss": 5.3735,
|
|
"mean_token_accuracy": 0.2005894288420677,
|
|
"num_tokens": 11577383.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 6.159556818008423,
|
|
"epoch": 5.880159124813526,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002447090615499,
|
|
"loss": 5.4231,
|
|
"mean_token_accuracy": 0.1978358045220375,
|
|
"num_tokens": 11586899.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 6.131319570541382,
|
|
"epoch": 5.885131775236201,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024432213311057383,
|
|
"loss": 5.4138,
|
|
"mean_token_accuracy": 0.1965557873249054,
|
|
"num_tokens": 11596884.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 6.131331586837769,
|
|
"epoch": 5.890104425658876,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024393529709160544,
|
|
"loss": 5.328,
|
|
"mean_token_accuracy": 0.20765064805746078,
|
|
"num_tokens": 11606603.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 6.101806163787842,
|
|
"epoch": 5.895077076081551,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024354855465837967,
|
|
"loss": 5.3714,
|
|
"mean_token_accuracy": 0.2024843916296959,
|
|
"num_tokens": 11616202.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 6.1064536571502686,
|
|
"epoch": 5.900049726504227,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024316190697599955,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.2185174897313118,
|
|
"num_tokens": 11625054.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 6.131908845901489,
|
|
"epoch": 5.905022376926902,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024277535520928257,
|
|
"loss": 5.3695,
|
|
"mean_token_accuracy": 0.20276850759983062,
|
|
"num_tokens": 11634636.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 6.1501116275787355,
|
|
"epoch": 5.909995027349577,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00024238890052275742,
|
|
"loss": 5.426,
|
|
"mean_token_accuracy": 0.19846278578042983,
|
|
"num_tokens": 11643949.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 6.1639384746551515,
|
|
"epoch": 5.914967677772252,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024200254408066014,
|
|
"loss": 5.3847,
|
|
"mean_token_accuracy": 0.20324333012104034,
|
|
"num_tokens": 11653964.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 6.121033573150635,
|
|
"epoch": 5.919940328194928,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002416162870469308,
|
|
"loss": 5.375,
|
|
"mean_token_accuracy": 0.2085279032588005,
|
|
"num_tokens": 11662774.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 6.145546293258667,
|
|
"epoch": 5.924912978617603,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002412301305852102,
|
|
"loss": 5.4235,
|
|
"mean_token_accuracy": 0.19470224529504776,
|
|
"num_tokens": 11672504.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 6.161111545562744,
|
|
"epoch": 5.929885629040278,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024084407585883588,
|
|
"loss": 5.4395,
|
|
"mean_token_accuracy": 0.19870424419641494,
|
|
"num_tokens": 11682225.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 6.142094469070434,
|
|
"epoch": 5.934858279462953,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002404581240308391,
|
|
"loss": 5.4294,
|
|
"mean_token_accuracy": 0.1935520887374878,
|
|
"num_tokens": 11691886.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 6.107822513580322,
|
|
"epoch": 5.939830929885629,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000240072276263941,
|
|
"loss": 5.3207,
|
|
"mean_token_accuracy": 0.20431501567363738,
|
|
"num_tokens": 11701366.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 6.114920520782471,
|
|
"epoch": 5.944803580308305,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023968653372054933,
|
|
"loss": 5.2383,
|
|
"mean_token_accuracy": 0.21297746300697326,
|
|
"num_tokens": 11710309.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 6.18134503364563,
|
|
"epoch": 5.949776230730979,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00023930089756275474,
|
|
"loss": 5.4119,
|
|
"mean_token_accuracy": 0.19657336175441742,
|
|
"num_tokens": 11719123.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 6.081782341003418,
|
|
"epoch": 5.954748881153655,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002389153689523275,
|
|
"loss": 5.3546,
|
|
"mean_token_accuracy": 0.20628882497549056,
|
|
"num_tokens": 11728931.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 6.139286661148072,
|
|
"epoch": 5.9597215315763306,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023852994905071385,
|
|
"loss": 5.3684,
|
|
"mean_token_accuracy": 0.20012889951467513,
|
|
"num_tokens": 11738666.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 6.119302272796631,
|
|
"epoch": 5.964694181999006,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023814463901903238,
|
|
"loss": 5.3938,
|
|
"mean_token_accuracy": 0.20299657583236694,
|
|
"num_tokens": 11748530.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.964694181999006,
|
|
"eval_entropy": 5.984832518661077,
|
|
"eval_loss": 6.436122417449951,
|
|
"eval_mean_token_accuracy": 0.15077858897192137,
|
|
"eval_num_tokens": 11748530.0,
|
|
"eval_runtime": 1.9274,
|
|
"eval_samples_per_second": 1797.746,
|
|
"eval_steps_per_second": 225.172,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 6.1411027908325195,
|
|
"epoch": 5.969666832421681,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023775944001807077,
|
|
"loss": 5.3974,
|
|
"mean_token_accuracy": 0.19787939190864562,
|
|
"num_tokens": 11759611.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 6.18923397064209,
|
|
"epoch": 5.974639482844356,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023737435320828244,
|
|
"loss": 5.3493,
|
|
"mean_token_accuracy": 0.20397916287183762,
|
|
"num_tokens": 11769427.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 6.046359491348267,
|
|
"epoch": 5.979612133267032,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023698937974978257,
|
|
"loss": 5.2024,
|
|
"mean_token_accuracy": 0.21326684057712555,
|
|
"num_tokens": 11778992.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 6.023154163360596,
|
|
"epoch": 5.984584783689707,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023660452080234473,
|
|
"loss": 5.2573,
|
|
"mean_token_accuracy": 0.21880466789007186,
|
|
"num_tokens": 11789844.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 6.153777122497559,
|
|
"epoch": 5.989557434112382,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023621977752539785,
|
|
"loss": 5.3999,
|
|
"mean_token_accuracy": 0.20292673110961915,
|
|
"num_tokens": 11800336.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 6.132955026626587,
|
|
"epoch": 5.994530084535057,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002358351510780223,
|
|
"loss": 5.4041,
|
|
"mean_token_accuracy": 0.20367093086242677,
|
|
"num_tokens": 11810081.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 6.126826906204224,
|
|
"epoch": 5.999502734957733,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023545064261894623,
|
|
"loss": 5.375,
|
|
"mean_token_accuracy": 0.1987760081887245,
|
|
"num_tokens": 11820046.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 6.1063863966200085,
|
|
"epoch": 6.00397812033814,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023506625330654263,
|
|
"loss": 5.1315,
|
|
"mean_token_accuracy": 0.22761427031622994,
|
|
"num_tokens": 11828769.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 6.171661043167115,
|
|
"epoch": 6.008950770760816,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023468198429882548,
|
|
"loss": 5.1271,
|
|
"mean_token_accuracy": 0.2184654101729393,
|
|
"num_tokens": 11838217.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 6.166880512237549,
|
|
"epoch": 6.013923421183491,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002342978367534462,
|
|
"loss": 5.2583,
|
|
"mean_token_accuracy": 0.20920509248971939,
|
|
"num_tokens": 11847268.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 6.150281667709351,
|
|
"epoch": 6.018896071606166,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002339138118276905,
|
|
"loss": 5.2514,
|
|
"mean_token_accuracy": 0.2133314073085785,
|
|
"num_tokens": 11857781.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 6.10182557106018,
|
|
"epoch": 6.023868722028841,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023352991067847456,
|
|
"loss": 5.0823,
|
|
"mean_token_accuracy": 0.22362101078033447,
|
|
"num_tokens": 11867810.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 6.0501185894012455,
|
|
"epoch": 6.028841372451517,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023314613446234162,
|
|
"loss": 5.1194,
|
|
"mean_token_accuracy": 0.22473154813051224,
|
|
"num_tokens": 11877698.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 6.050595140457153,
|
|
"epoch": 6.033814022874192,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023276248433545867,
|
|
"loss": 5.2368,
|
|
"mean_token_accuracy": 0.21160527914762498,
|
|
"num_tokens": 11888696.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 6.194988775253296,
|
|
"epoch": 6.038786673296867,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023237896145361276,
|
|
"loss": 5.2841,
|
|
"mean_token_accuracy": 0.2105478972196579,
|
|
"num_tokens": 11898865.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 6.113224744796753,
|
|
"epoch": 6.043759323719542,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023199556697220758,
|
|
"loss": 5.1297,
|
|
"mean_token_accuracy": 0.2266096979379654,
|
|
"num_tokens": 11909506.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 6.100506448745728,
|
|
"epoch": 6.048731974142218,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023161230204626015,
|
|
"loss": 5.1573,
|
|
"mean_token_accuracy": 0.2178516685962677,
|
|
"num_tokens": 11918663.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 6.198505640029907,
|
|
"epoch": 6.053704624564893,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023122916783039694,
|
|
"loss": 5.2868,
|
|
"mean_token_accuracy": 0.21335690915584565,
|
|
"num_tokens": 11928683.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 6.10963134765625,
|
|
"epoch": 6.058677274987568,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002308461654788509,
|
|
"loss": 5.1181,
|
|
"mean_token_accuracy": 0.22794189453125,
|
|
"num_tokens": 11937251.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 6.120733642578125,
|
|
"epoch": 6.063649925410243,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00023046329614545758,
|
|
"loss": 5.1392,
|
|
"mean_token_accuracy": 0.21992189884185792,
|
|
"num_tokens": 11946061.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 6.082307004928589,
|
|
"epoch": 6.068622575832919,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023008056098365183,
|
|
"loss": 5.1204,
|
|
"mean_token_accuracy": 0.21948319673538208,
|
|
"num_tokens": 11956512.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 6.087352895736695,
|
|
"epoch": 6.073595226255594,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022969796114646424,
|
|
"loss": 5.2281,
|
|
"mean_token_accuracy": 0.21192947775125504,
|
|
"num_tokens": 11966052.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 6.13516755104065,
|
|
"epoch": 6.078567876678269,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022931549778651783,
|
|
"loss": 5.1588,
|
|
"mean_token_accuracy": 0.22389823347330093,
|
|
"num_tokens": 11974742.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 6.130410003662109,
|
|
"epoch": 6.083540527100944,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022893317205602444,
|
|
"loss": 5.1557,
|
|
"mean_token_accuracy": 0.22812108546495438,
|
|
"num_tokens": 11985009.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 6.092312002182007,
|
|
"epoch": 6.08851317752362,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022855098510678125,
|
|
"loss": 5.2266,
|
|
"mean_token_accuracy": 0.21740840822458268,
|
|
"num_tokens": 11994685.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 6.129199743270874,
|
|
"epoch": 6.093485827946296,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00022816893809016737,
|
|
"loss": 5.2425,
|
|
"mean_token_accuracy": 0.2101394861936569,
|
|
"num_tokens": 12004225.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 6.052749490737915,
|
|
"epoch": 6.098458478368971,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022778703215714014,
|
|
"loss": 5.163,
|
|
"mean_token_accuracy": 0.21831387281417847,
|
|
"num_tokens": 12013422.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.963638639450073,
|
|
"epoch": 6.103431128791646,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00022740526845823223,
|
|
"loss": 5.0295,
|
|
"mean_token_accuracy": 0.2357778713107109,
|
|
"num_tokens": 12023739.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 6.17730712890625,
|
|
"epoch": 6.1084037792143215,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002270236481435477,
|
|
"loss": 5.2786,
|
|
"mean_token_accuracy": 0.20921764969825746,
|
|
"num_tokens": 12034189.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 6.194540405273438,
|
|
"epoch": 6.113376429636997,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022664217236275847,
|
|
"loss": 5.276,
|
|
"mean_token_accuracy": 0.2070918411016464,
|
|
"num_tokens": 12043134.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 6.134635639190674,
|
|
"epoch": 6.118349080059672,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022626084226510124,
|
|
"loss": 5.1498,
|
|
"mean_token_accuracy": 0.21717145591974257,
|
|
"num_tokens": 12052843.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 6.129702138900757,
|
|
"epoch": 6.123321730482347,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022587965899937364,
|
|
"loss": 5.2209,
|
|
"mean_token_accuracy": 0.21413696110248565,
|
|
"num_tokens": 12063260.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 6.131983375549316,
|
|
"epoch": 6.1282943809050225,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002254986237139312,
|
|
"loss": 5.2322,
|
|
"mean_token_accuracy": 0.21542789041996002,
|
|
"num_tokens": 12072534.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 6.168384218215943,
|
|
"epoch": 6.133267031327698,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022511773755668342,
|
|
"loss": 5.3228,
|
|
"mean_token_accuracy": 0.20404146313667298,
|
|
"num_tokens": 12081796.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.989533185958862,
|
|
"epoch": 6.138239681750373,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002247370016750907,
|
|
"loss": 5.0716,
|
|
"mean_token_accuracy": 0.2284193977713585,
|
|
"num_tokens": 12092885.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 6.1171746253967285,
|
|
"epoch": 6.143212332173048,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002243564172161605,
|
|
"loss": 5.1965,
|
|
"mean_token_accuracy": 0.2156907305121422,
|
|
"num_tokens": 12102432.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.971362972259522,
|
|
"epoch": 6.1481849825957235,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022397598532644441,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.2316624939441681,
|
|
"num_tokens": 12112245.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 6.143703937530518,
|
|
"epoch": 6.153157633018399,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022359570715203408,
|
|
"loss": 5.1374,
|
|
"mean_token_accuracy": 0.21712316274642945,
|
|
"num_tokens": 12121850.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 6.097066116333008,
|
|
"epoch": 6.158130283441074,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022321558383855828,
|
|
"loss": 5.0721,
|
|
"mean_token_accuracy": 0.22947365194559097,
|
|
"num_tokens": 12132538.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.926192903518677,
|
|
"epoch": 6.163102933863749,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002228356165311791,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.23892980366945266,
|
|
"num_tokens": 12142041.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 6.152984380722046,
|
|
"epoch": 6.1680755842864246,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022245580637458883,
|
|
"loss": 5.3225,
|
|
"mean_token_accuracy": 0.208046256005764,
|
|
"num_tokens": 12151075.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 6.2084297180175785,
|
|
"epoch": 6.1730482347091,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022207615451300616,
|
|
"loss": 5.2393,
|
|
"mean_token_accuracy": 0.21233527660369872,
|
|
"num_tokens": 12160657.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 6.192020130157471,
|
|
"epoch": 6.178020885131775,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022169666209017304,
|
|
"loss": 5.3551,
|
|
"mean_token_accuracy": 0.20679855197668076,
|
|
"num_tokens": 12171692.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.989048862457276,
|
|
"epoch": 6.18299353555445,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002213173302493509,
|
|
"loss": 5.0906,
|
|
"mean_token_accuracy": 0.22451018840074538,
|
|
"num_tokens": 12181429.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 6.046659469604492,
|
|
"epoch": 6.1879661859771256,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022093816013331763,
|
|
"loss": 5.1554,
|
|
"mean_token_accuracy": 0.22534767538309097,
|
|
"num_tokens": 12191992.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.988919925689697,
|
|
"epoch": 6.192938836399801,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00022055915288436362,
|
|
"loss": 5.13,
|
|
"mean_token_accuracy": 0.22963055670261384,
|
|
"num_tokens": 12200804.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 6.23975772857666,
|
|
"epoch": 6.197911486822476,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022018030964428897,
|
|
"loss": 5.2833,
|
|
"mean_token_accuracy": 0.20633082687854767,
|
|
"num_tokens": 12211774.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 6.113170719146728,
|
|
"epoch": 6.202884137245151,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021980163155439943,
|
|
"loss": 5.1761,
|
|
"mean_token_accuracy": 0.21701433956623079,
|
|
"num_tokens": 12220270.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 6.10855541229248,
|
|
"epoch": 6.207856787667827,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021942311975550323,
|
|
"loss": 5.2489,
|
|
"mean_token_accuracy": 0.21313494443893433,
|
|
"num_tokens": 12230662.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 6.118883848190308,
|
|
"epoch": 6.212829438090502,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021904477538790774,
|
|
"loss": 5.3297,
|
|
"mean_token_accuracy": 0.2072910711169243,
|
|
"num_tokens": 12240383.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 6.131918811798096,
|
|
"epoch": 6.217802088513178,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00021866659959141578,
|
|
"loss": 5.2777,
|
|
"mean_token_accuracy": 0.21289461702108384,
|
|
"num_tokens": 12250247.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 6.001589727401734,
|
|
"epoch": 6.222774738935853,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002182885935053225,
|
|
"loss": 5.1195,
|
|
"mean_token_accuracy": 0.2317013531923294,
|
|
"num_tokens": 12261245.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 6.1774393081665036,
|
|
"epoch": 6.2277473893585285,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021791075826841166,
|
|
"loss": 5.2899,
|
|
"mean_token_accuracy": 0.20467288047075272,
|
|
"num_tokens": 12271596.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 6.243044376373291,
|
|
"epoch": 6.232720039781204,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002175330950189523,
|
|
"loss": 5.3543,
|
|
"mean_token_accuracy": 0.2060009479522705,
|
|
"num_tokens": 12283124.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 6.080815553665161,
|
|
"epoch": 6.237692690203879,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002171556048946955,
|
|
"loss": 5.1784,
|
|
"mean_token_accuracy": 0.220473712682724,
|
|
"num_tokens": 12293399.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.908518075942993,
|
|
"epoch": 6.242665340626554,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021677828903287063,
|
|
"loss": 5.0673,
|
|
"mean_token_accuracy": 0.2351531445980072,
|
|
"num_tokens": 12303396.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 6.1490448951721195,
|
|
"epoch": 6.2476379910492295,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021640114857018205,
|
|
"loss": 5.2177,
|
|
"mean_token_accuracy": 0.21578546166419982,
|
|
"num_tokens": 12312483.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 6.089865875244141,
|
|
"epoch": 6.252610641471905,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021602418464280586,
|
|
"loss": 5.1052,
|
|
"mean_token_accuracy": 0.22903864532709123,
|
|
"num_tokens": 12322420.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 6.260043907165527,
|
|
"epoch": 6.25758329189458,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021564739838638615,
|
|
"loss": 5.4047,
|
|
"mean_token_accuracy": 0.1984893798828125,
|
|
"num_tokens": 12331740.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 6.15282301902771,
|
|
"epoch": 6.262555942317255,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021527079093603202,
|
|
"loss": 5.2803,
|
|
"mean_token_accuracy": 0.20585861206054687,
|
|
"num_tokens": 12341803.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 6.110366535186768,
|
|
"epoch": 6.2675285927399305,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021489436342631352,
|
|
"loss": 5.1813,
|
|
"mean_token_accuracy": 0.21318930387496948,
|
|
"num_tokens": 12350824.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 6.142373943328858,
|
|
"epoch": 6.272501243162606,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021451811699125883,
|
|
"loss": 5.2313,
|
|
"mean_token_accuracy": 0.2076294019818306,
|
|
"num_tokens": 12360116.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 6.112145328521729,
|
|
"epoch": 6.277473893585281,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021414205276435073,
|
|
"loss": 5.2218,
|
|
"mean_token_accuracy": 0.2074524089694023,
|
|
"num_tokens": 12370486.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 6.085908603668213,
|
|
"epoch": 6.282446544007956,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002137661718785229,
|
|
"loss": 5.1163,
|
|
"mean_token_accuracy": 0.2188025951385498,
|
|
"num_tokens": 12380178.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 6.112122583389282,
|
|
"epoch": 6.2874191944306315,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00021339047546615675,
|
|
"loss": 5.2604,
|
|
"mean_token_accuracy": 0.20549062490463257,
|
|
"num_tokens": 12389844.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 6.1463623046875,
|
|
"epoch": 6.292391844853307,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00021301496465907794,
|
|
"loss": 5.2726,
|
|
"mean_token_accuracy": 0.21172718554735184,
|
|
"num_tokens": 12400880.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 6.1779014587402346,
|
|
"epoch": 6.297364495275982,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021263964058855294,
|
|
"loss": 5.2577,
|
|
"mean_token_accuracy": 0.21173717677593232,
|
|
"num_tokens": 12410833.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 6.139164733886719,
|
|
"epoch": 6.302337145698657,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00021226450438528572,
|
|
"loss": 5.1723,
|
|
"mean_token_accuracy": 0.2282631054520607,
|
|
"num_tokens": 12419993.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 6.115909481048584,
|
|
"epoch": 6.3073097961213325,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00021188955717941428,
|
|
"loss": 5.2892,
|
|
"mean_token_accuracy": 0.2149337574839592,
|
|
"num_tokens": 12431395.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 6.039453411102295,
|
|
"epoch": 6.312282446544008,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002115148001005071,
|
|
"loss": 5.1184,
|
|
"mean_token_accuracy": 0.21913209855556487,
|
|
"num_tokens": 12440961.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 6.141243600845337,
|
|
"epoch": 6.317255096966683,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021114023427756012,
|
|
"loss": 5.2885,
|
|
"mean_token_accuracy": 0.21156540811061858,
|
|
"num_tokens": 12450333.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 6.154447221755982,
|
|
"epoch": 6.322227747389358,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021076586083899296,
|
|
"loss": 5.2816,
|
|
"mean_token_accuracy": 0.20830729156732558,
|
|
"num_tokens": 12459373.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 6.047233581542969,
|
|
"epoch": 6.3272003978120335,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002103916809126456,
|
|
"loss": 5.1526,
|
|
"mean_token_accuracy": 0.22213816791772842,
|
|
"num_tokens": 12469366.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 6.171606636047363,
|
|
"epoch": 6.332173048234709,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00021001769562577517,
|
|
"loss": 5.2097,
|
|
"mean_token_accuracy": 0.215470153093338,
|
|
"num_tokens": 12478650.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.88722243309021,
|
|
"epoch": 6.337145698657384,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002096439061050523,
|
|
"loss": 4.9206,
|
|
"mean_token_accuracy": 0.2450374811887741,
|
|
"num_tokens": 12490607.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 6.070699834823609,
|
|
"epoch": 6.34211834908006,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000209270313476558,
|
|
"loss": 5.262,
|
|
"mean_token_accuracy": 0.21041743606328964,
|
|
"num_tokens": 12502893.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.953101873397827,
|
|
"epoch": 6.347090999502735,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020889691886578004,
|
|
"loss": 5.0389,
|
|
"mean_token_accuracy": 0.239610493183136,
|
|
"num_tokens": 12513470.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 6.159112596511841,
|
|
"epoch": 6.352063649925411,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020852372339760973,
|
|
"loss": 5.2533,
|
|
"mean_token_accuracy": 0.21283662170171738,
|
|
"num_tokens": 12522982.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 6.015686416625977,
|
|
"epoch": 6.357036300348086,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002081507281963383,
|
|
"loss": 5.1546,
|
|
"mean_token_accuracy": 0.22372486144304277,
|
|
"num_tokens": 12533247.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 6.0238159656524655,
|
|
"epoch": 6.362008950770761,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020777793438565368,
|
|
"loss": 5.2904,
|
|
"mean_token_accuracy": 0.2094276025891304,
|
|
"num_tokens": 12543488.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 6.159257411956787,
|
|
"epoch": 6.366981601193436,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020740534308863717,
|
|
"loss": 5.3279,
|
|
"mean_token_accuracy": 0.21000277400016784,
|
|
"num_tokens": 12553782.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 6.158183240890503,
|
|
"epoch": 6.371954251616112,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00020703295542775992,
|
|
"loss": 5.3568,
|
|
"mean_token_accuracy": 0.20297116041183472,
|
|
"num_tokens": 12562913.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 6.095121908187866,
|
|
"epoch": 6.376926902038787,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020666077252487968,
|
|
"loss": 5.1272,
|
|
"mean_token_accuracy": 0.22526115030050278,
|
|
"num_tokens": 12571979.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.875943231582641,
|
|
"epoch": 6.381899552461462,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002062887955012372,
|
|
"loss": 4.9651,
|
|
"mean_token_accuracy": 0.2469969943165779,
|
|
"num_tokens": 12581445.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 6.149869441986084,
|
|
"epoch": 6.386872202884137,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020591702547745306,
|
|
"loss": 5.3321,
|
|
"mean_token_accuracy": 0.20134178400039673,
|
|
"num_tokens": 12592437.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 6.092759275436402,
|
|
"epoch": 6.391844853306813,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020554546357352433,
|
|
"loss": 5.2064,
|
|
"mean_token_accuracy": 0.22167444080114365,
|
|
"num_tokens": 12602946.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 6.185548448562622,
|
|
"epoch": 6.396817503729488,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000205174110908821,
|
|
"loss": 5.289,
|
|
"mean_token_accuracy": 0.2086523786187172,
|
|
"num_tokens": 12612620.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 6.11659426689148,
|
|
"epoch": 6.401790154152163,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002048029686020827,
|
|
"loss": 5.2222,
|
|
"mean_token_accuracy": 0.21012657582759858,
|
|
"num_tokens": 12621588.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 6.055890655517578,
|
|
"epoch": 6.406762804574838,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002044320377714154,
|
|
"loss": 5.1418,
|
|
"mean_token_accuracy": 0.21970828771591186,
|
|
"num_tokens": 12630802.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 6.073174905776978,
|
|
"epoch": 6.411735454997514,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020406131953428785,
|
|
"loss": 5.1477,
|
|
"mean_token_accuracy": 0.2173631340265274,
|
|
"num_tokens": 12640358.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 6.127038955688477,
|
|
"epoch": 6.416708105420189,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002036908150075285,
|
|
"loss": 5.3298,
|
|
"mean_token_accuracy": 0.20495595335960387,
|
|
"num_tokens": 12650598.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 6.107139730453492,
|
|
"epoch": 6.421680755842864,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002033205253073218,
|
|
"loss": 5.1855,
|
|
"mean_token_accuracy": 0.21921363472938538,
|
|
"num_tokens": 12661213.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 6.1991557598114015,
|
|
"epoch": 6.426653406265539,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020295045154920528,
|
|
"loss": 5.2265,
|
|
"mean_token_accuracy": 0.2186043679714203,
|
|
"num_tokens": 12670374.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 6.085563278198242,
|
|
"epoch": 6.431626056688215,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020258059484806564,
|
|
"loss": 5.2164,
|
|
"mean_token_accuracy": 0.21671980023384094,
|
|
"num_tokens": 12679725.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 6.029938554763794,
|
|
"epoch": 6.43659870711089,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020221095631813574,
|
|
"loss": 5.1848,
|
|
"mean_token_accuracy": 0.21273670494556426,
|
|
"num_tokens": 12688391.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 6.128770017623902,
|
|
"epoch": 6.441571357533565,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002018415370729913,
|
|
"loss": 5.2854,
|
|
"mean_token_accuracy": 0.21072024255990982,
|
|
"num_tokens": 12698030.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 6.1133880615234375,
|
|
"epoch": 6.44654400795624,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00020147233822554722,
|
|
"loss": 5.2858,
|
|
"mean_token_accuracy": 0.21788574159145355,
|
|
"num_tokens": 12708577.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 6.192506599426269,
|
|
"epoch": 6.451516658378916,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020110336088805463,
|
|
"loss": 5.2883,
|
|
"mean_token_accuracy": 0.20656535625457764,
|
|
"num_tokens": 12718573.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 6.143504095077515,
|
|
"epoch": 6.456489308801591,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020073460617209732,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.21594239324331282,
|
|
"num_tokens": 12728246.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 6.045164966583252,
|
|
"epoch": 6.461461959224266,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00020036607518858824,
|
|
"loss": 5.2069,
|
|
"mean_token_accuracy": 0.2203689232468605,
|
|
"num_tokens": 12737807.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 6.145089721679687,
|
|
"epoch": 6.466434609646942,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019999776904776646,
|
|
"loss": 5.3135,
|
|
"mean_token_accuracy": 0.20938613563776015,
|
|
"num_tokens": 12747463.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 6.1644655704498295,
|
|
"epoch": 6.471407260069617,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001996296888591937,
|
|
"loss": 5.2343,
|
|
"mean_token_accuracy": 0.20752496272325516,
|
|
"num_tokens": 12757449.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 6.155842971801758,
|
|
"epoch": 6.476379910492293,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019926183573175084,
|
|
"loss": 5.1871,
|
|
"mean_token_accuracy": 0.21913132220506668,
|
|
"num_tokens": 12767698.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 6.05535774230957,
|
|
"epoch": 6.481352560914968,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001988942107736348,
|
|
"loss": 5.1958,
|
|
"mean_token_accuracy": 0.209917850792408,
|
|
"num_tokens": 12776252.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 6.069735908508301,
|
|
"epoch": 6.486325211337643,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019852681509235516,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.22044609636068344,
|
|
"num_tokens": 12785001.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 6.10723180770874,
|
|
"epoch": 6.4912978617603185,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019815964979473088,
|
|
"loss": 5.2618,
|
|
"mean_token_accuracy": 0.20891633927822112,
|
|
"num_tokens": 12794587.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 6.132402753829956,
|
|
"epoch": 6.496270512182994,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019779271598688642,
|
|
"loss": 5.2174,
|
|
"mean_token_accuracy": 0.21521003246307374,
|
|
"num_tokens": 12803439.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 6.114540100097656,
|
|
"epoch": 6.501243162605669,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019742601477424937,
|
|
"loss": 5.2595,
|
|
"mean_token_accuracy": 0.20553765892982484,
|
|
"num_tokens": 12813039.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 6.067819213867187,
|
|
"epoch": 6.506215813028344,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001970595472615463,
|
|
"loss": 5.2022,
|
|
"mean_token_accuracy": 0.21588290482759476,
|
|
"num_tokens": 12823105.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 6.026543378829956,
|
|
"epoch": 6.5111884634510195,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019669331455279989,
|
|
"loss": 5.1858,
|
|
"mean_token_accuracy": 0.21988351941108703,
|
|
"num_tokens": 12833576.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 6.121774435043335,
|
|
"epoch": 6.516161113873695,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019632731775132522,
|
|
"loss": 5.3319,
|
|
"mean_token_accuracy": 0.20588493645191192,
|
|
"num_tokens": 12843570.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 6.151248264312744,
|
|
"epoch": 6.52113376429637,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00019596155795972698,
|
|
"loss": 5.3009,
|
|
"mean_token_accuracy": 0.21359962970018387,
|
|
"num_tokens": 12852751.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 6.140508604049683,
|
|
"epoch": 6.526106414719045,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001955960362798957,
|
|
"loss": 5.2507,
|
|
"mean_token_accuracy": 0.21075770556926726,
|
|
"num_tokens": 12863114.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 6.086218166351318,
|
|
"epoch": 6.5310790651417205,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001952307538130045,
|
|
"loss": 5.1251,
|
|
"mean_token_accuracy": 0.21657571643590928,
|
|
"num_tokens": 12871731.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 6.053445339202881,
|
|
"epoch": 6.536051715564396,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001948657116595059,
|
|
"loss": 5.2145,
|
|
"mean_token_accuracy": 0.2109370708465576,
|
|
"num_tokens": 12880261.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 6.0765081405639645,
|
|
"epoch": 6.541024365987071,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019450091091912847,
|
|
"loss": 5.1517,
|
|
"mean_token_accuracy": 0.22010283470153807,
|
|
"num_tokens": 12889528.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.9566949844360355,
|
|
"epoch": 6.545997016409746,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001941363526908736,
|
|
"loss": 5.1038,
|
|
"mean_token_accuracy": 0.23381876498460769,
|
|
"num_tokens": 12899734.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 6.132054996490479,
|
|
"epoch": 6.5509696668324215,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00019377203807301185,
|
|
"loss": 5.1991,
|
|
"mean_token_accuracy": 0.21433150470256807,
|
|
"num_tokens": 12909960.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 6.093586874008179,
|
|
"epoch": 6.555942317255097,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019340796816308,
|
|
"loss": 5.2081,
|
|
"mean_token_accuracy": 0.21909433901309966,
|
|
"num_tokens": 12919411.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 6.133450984954834,
|
|
"epoch": 6.560914967677772,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019304414405787768,
|
|
"loss": 5.2983,
|
|
"mean_token_accuracy": 0.2152979075908661,
|
|
"num_tokens": 12929470.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 6.102458143234253,
|
|
"epoch": 6.565887618100447,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001926805668534639,
|
|
"loss": 5.2351,
|
|
"mean_token_accuracy": 0.21348885148763658,
|
|
"num_tokens": 12940292.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 6.111742401123047,
|
|
"epoch": 6.5708602685231225,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019231723764515397,
|
|
"loss": 5.2553,
|
|
"mean_token_accuracy": 0.20518381297588348,
|
|
"num_tokens": 12950218.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 6.128354454040528,
|
|
"epoch": 6.575832918945798,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00019195415752751595,
|
|
"loss": 5.32,
|
|
"mean_token_accuracy": 0.2076724201440811,
|
|
"num_tokens": 12961082.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 6.138163232803345,
|
|
"epoch": 6.580805569368473,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00019159132759436764,
|
|
"loss": 5.3017,
|
|
"mean_token_accuracy": 0.20447692573070525,
|
|
"num_tokens": 12970913.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 6.180662679672241,
|
|
"epoch": 6.585778219791148,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019122874893877294,
|
|
"loss": 5.3037,
|
|
"mean_token_accuracy": 0.21025364995002746,
|
|
"num_tokens": 12981201.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 6.152100658416748,
|
|
"epoch": 6.590750870213824,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019086642265303894,
|
|
"loss": 5.3106,
|
|
"mean_token_accuracy": 0.20645617097616195,
|
|
"num_tokens": 12991283.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 6.082708692550659,
|
|
"epoch": 6.595723520636499,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001905043498287124,
|
|
"loss": 5.1146,
|
|
"mean_token_accuracy": 0.23013215959072114,
|
|
"num_tokens": 13001201.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.93934178352356,
|
|
"epoch": 6.600696171059175,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001901425315565764,
|
|
"loss": 5.033,
|
|
"mean_token_accuracy": 0.23993124961853027,
|
|
"num_tokens": 13011119.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 6.1316615581512455,
|
|
"epoch": 6.60566882148185,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018978096892664726,
|
|
"loss": 5.2234,
|
|
"mean_token_accuracy": 0.21717022359371185,
|
|
"num_tokens": 13021459.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 6.003587341308593,
|
|
"epoch": 6.6106414719045254,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000189419663028171,
|
|
"loss": 5.1446,
|
|
"mean_token_accuracy": 0.2219623312354088,
|
|
"num_tokens": 13032021.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 6.095711946487427,
|
|
"epoch": 6.615614122327201,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00018905861494962034,
|
|
"loss": 5.267,
|
|
"mean_token_accuracy": 0.21032191812992096,
|
|
"num_tokens": 13041418.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 6.141036796569824,
|
|
"epoch": 6.620586772749876,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018869782577869132,
|
|
"loss": 5.2544,
|
|
"mean_token_accuracy": 0.21142735481262206,
|
|
"num_tokens": 13051535.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 6.146865129470825,
|
|
"epoch": 6.625559423172551,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018833729660229997,
|
|
"loss": 5.1853,
|
|
"mean_token_accuracy": 0.21481041610240936,
|
|
"num_tokens": 13060854.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 6.117269468307495,
|
|
"epoch": 6.6305320735952264,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00018797702850657889,
|
|
"loss": 5.2295,
|
|
"mean_token_accuracy": 0.21513847559690474,
|
|
"num_tokens": 13069429.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 6.091485261917114,
|
|
"epoch": 6.635504724017902,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001876170225768744,
|
|
"loss": 5.2153,
|
|
"mean_token_accuracy": 0.21897751986980438,
|
|
"num_tokens": 13078990.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 6.00847373008728,
|
|
"epoch": 6.640477374440577,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001872572798977429,
|
|
"loss": 5.1191,
|
|
"mean_token_accuracy": 0.23233832269906998,
|
|
"num_tokens": 13089702.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 6.1030079364776615,
|
|
"epoch": 6.645450024863252,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001868978015529476,
|
|
"loss": 5.208,
|
|
"mean_token_accuracy": 0.2099670797586441,
|
|
"num_tokens": 13099890.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 6.1266028881073,
|
|
"epoch": 6.6504226752859275,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018653858862545565,
|
|
"loss": 5.309,
|
|
"mean_token_accuracy": 0.20586056262254715,
|
|
"num_tokens": 13108767.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.922989177703857,
|
|
"epoch": 6.655395325708603,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018617964219743428,
|
|
"loss": 4.9145,
|
|
"mean_token_accuracy": 0.24034570902585983,
|
|
"num_tokens": 13118548.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 6.068158578872681,
|
|
"epoch": 6.660367976131278,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018582096335024813,
|
|
"loss": 5.2628,
|
|
"mean_token_accuracy": 0.21723951846361161,
|
|
"num_tokens": 13129451.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 6.0600457191467285,
|
|
"epoch": 6.665340626553953,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001854625531644557,
|
|
"loss": 5.2642,
|
|
"mean_token_accuracy": 0.20641745179891585,
|
|
"num_tokens": 13138679.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 6.139621305465698,
|
|
"epoch": 6.6703132769766285,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018510441271980588,
|
|
"loss": 5.3413,
|
|
"mean_token_accuracy": 0.1990756630897522,
|
|
"num_tokens": 13148980.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 6.100678539276123,
|
|
"epoch": 6.675285927399304,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018474654309523508,
|
|
"loss": 5.1614,
|
|
"mean_token_accuracy": 0.22345857322216034,
|
|
"num_tokens": 13158743.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 6.077292013168335,
|
|
"epoch": 6.680258577821979,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00018438894536886396,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.22114323526620866,
|
|
"num_tokens": 13168371.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 6.11578598022461,
|
|
"epoch": 6.685231228244654,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00018403162061799395,
|
|
"loss": 5.2343,
|
|
"mean_token_accuracy": 0.21323888301849364,
|
|
"num_tokens": 13177399.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 6.100509548187256,
|
|
"epoch": 6.6902038786673295,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018367456991910403,
|
|
"loss": 5.1605,
|
|
"mean_token_accuracy": 0.21486492305994034,
|
|
"num_tokens": 13186689.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 6.1140604496002195,
|
|
"epoch": 6.695176529090005,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00018331779434784773,
|
|
"loss": 5.1649,
|
|
"mean_token_accuracy": 0.21425800621509553,
|
|
"num_tokens": 13198859.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 6.146596479415893,
|
|
"epoch": 6.70014917951268,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018296129497904955,
|
|
"loss": 5.1969,
|
|
"mean_token_accuracy": 0.21019037067890167,
|
|
"num_tokens": 13207237.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 6.108058929443359,
|
|
"epoch": 6.705121829935355,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001826050728867021,
|
|
"loss": 5.2271,
|
|
"mean_token_accuracy": 0.21340237706899642,
|
|
"num_tokens": 13216163.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 6.183069944381714,
|
|
"epoch": 6.7100944803580305,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00018224912914396253,
|
|
"loss": 5.3456,
|
|
"mean_token_accuracy": 0.20497251451015472,
|
|
"num_tokens": 13225714.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 6.060117149353028,
|
|
"epoch": 6.715067130780707,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018189346482314945,
|
|
"loss": 5.2153,
|
|
"mean_token_accuracy": 0.21829520761966706,
|
|
"num_tokens": 13234226.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 6.129011964797973,
|
|
"epoch": 6.720039781203381,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00018153808099573976,
|
|
"loss": 5.2998,
|
|
"mean_token_accuracy": 0.20473092049360275,
|
|
"num_tokens": 13244286.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 6.073872518539429,
|
|
"epoch": 6.725012431626057,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018118297873236517,
|
|
"loss": 5.2287,
|
|
"mean_token_accuracy": 0.21004527509212495,
|
|
"num_tokens": 13254283.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 6.104308795928955,
|
|
"epoch": 6.729985082048732,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018082815910280925,
|
|
"loss": 5.1897,
|
|
"mean_token_accuracy": 0.2142678827047348,
|
|
"num_tokens": 13263886.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 6.148085260391236,
|
|
"epoch": 6.734957732471408,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018047362317600424,
|
|
"loss": 5.2609,
|
|
"mean_token_accuracy": 0.21258271634578704,
|
|
"num_tokens": 13273406.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.986249542236328,
|
|
"epoch": 6.739930382894083,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00018011937202002748,
|
|
"loss": 5.108,
|
|
"mean_token_accuracy": 0.23981691747903824,
|
|
"num_tokens": 13282980.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 6.052365827560425,
|
|
"epoch": 6.744903033316758,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017976540670209856,
|
|
"loss": 5.1735,
|
|
"mean_token_accuracy": 0.21678052395582198,
|
|
"num_tokens": 13292791.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.9461534976959225,
|
|
"epoch": 6.749875683739433,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017941172828857572,
|
|
"loss": 5.069,
|
|
"mean_token_accuracy": 0.24287063330411912,
|
|
"num_tokens": 13302440.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 6.057154798507691,
|
|
"epoch": 6.754848334162109,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017905833784495325,
|
|
"loss": 5.213,
|
|
"mean_token_accuracy": 0.21266716569662095,
|
|
"num_tokens": 13312465.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 6.1581511974334715,
|
|
"epoch": 6.759820984584784,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017870523643585769,
|
|
"loss": 5.2319,
|
|
"mean_token_accuracy": 0.2146006405353546,
|
|
"num_tokens": 13321991.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 6.032607316970825,
|
|
"epoch": 6.764793635007459,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017835242512504463,
|
|
"loss": 5.1284,
|
|
"mean_token_accuracy": 0.22108317762613297,
|
|
"num_tokens": 13331671.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 6.093976831436157,
|
|
"epoch": 6.769766285430134,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017799990497539608,
|
|
"loss": 5.2728,
|
|
"mean_token_accuracy": 0.2080679088830948,
|
|
"num_tokens": 13341196.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 6.16185245513916,
|
|
"epoch": 6.77473893585281,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001776476770489167,
|
|
"loss": 5.2699,
|
|
"mean_token_accuracy": 0.20136482566595076,
|
|
"num_tokens": 13350993.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 6.137589168548584,
|
|
"epoch": 6.779711586275485,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017729574240673069,
|
|
"loss": 5.2129,
|
|
"mean_token_accuracy": 0.21477311998605728,
|
|
"num_tokens": 13360385.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 6.0825889110565186,
|
|
"epoch": 6.78468423669816,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017694410210907906,
|
|
"loss": 5.2355,
|
|
"mean_token_accuracy": 0.21925479024648667,
|
|
"num_tokens": 13370057.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.971081686019898,
|
|
"epoch": 6.789656887120835,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001765927572153156,
|
|
"loss": 5.0492,
|
|
"mean_token_accuracy": 0.2374481201171875,
|
|
"num_tokens": 13380379.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 6.151688528060913,
|
|
"epoch": 6.794629537543511,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001762417087839046,
|
|
"loss": 5.2973,
|
|
"mean_token_accuracy": 0.2081499069929123,
|
|
"num_tokens": 13389795.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 6.146138000488281,
|
|
"epoch": 6.799602187966186,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017589095787241688,
|
|
"loss": 5.2276,
|
|
"mean_token_accuracy": 0.21263906806707383,
|
|
"num_tokens": 13400630.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 6.130971527099609,
|
|
"epoch": 6.804574838388861,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001755405055375272,
|
|
"loss": 5.2187,
|
|
"mean_token_accuracy": 0.21258143633604049,
|
|
"num_tokens": 13411044.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 6.165171480178833,
|
|
"epoch": 6.809547488811536,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017519035283501084,
|
|
"loss": 5.2247,
|
|
"mean_token_accuracy": 0.20420544743537902,
|
|
"num_tokens": 13420092.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 6.087076663970947,
|
|
"epoch": 6.814520139234212,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017484050081974004,
|
|
"loss": 5.2487,
|
|
"mean_token_accuracy": 0.21925239264965057,
|
|
"num_tokens": 13430058.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 6.093574810028076,
|
|
"epoch": 6.819492789656887,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001744909505456817,
|
|
"loss": 5.234,
|
|
"mean_token_accuracy": 0.2136865496635437,
|
|
"num_tokens": 13439061.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 6.141365957260132,
|
|
"epoch": 6.824465440079562,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017414170306589318,
|
|
"loss": 5.2998,
|
|
"mean_token_accuracy": 0.21469437927007676,
|
|
"num_tokens": 13449341.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 6.163508892059326,
|
|
"epoch": 6.829438090502237,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001737927594325201,
|
|
"loss": 5.2594,
|
|
"mean_token_accuracy": 0.2137780413031578,
|
|
"num_tokens": 13457896.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 6.102454996109008,
|
|
"epoch": 6.834410740924913,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017344412069679257,
|
|
"loss": 5.2326,
|
|
"mean_token_accuracy": 0.21294221580028533,
|
|
"num_tokens": 13468404.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 6.125158977508545,
|
|
"epoch": 6.839383391347588,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001730957879090218,
|
|
"loss": 5.2766,
|
|
"mean_token_accuracy": 0.21129491180181503,
|
|
"num_tokens": 13477530.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 6.12853307723999,
|
|
"epoch": 6.844356041770263,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017274776211859778,
|
|
"loss": 5.3085,
|
|
"mean_token_accuracy": 0.20548388808965684,
|
|
"num_tokens": 13487250.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 6.116113090515137,
|
|
"epoch": 6.849328692192939,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017240004437398554,
|
|
"loss": 5.2933,
|
|
"mean_token_accuracy": 0.20591281950473786,
|
|
"num_tokens": 13497342.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 6.208527040481568,
|
|
"epoch": 6.854301342615614,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001720526357227218,
|
|
"loss": 5.3234,
|
|
"mean_token_accuracy": 0.2085000678896904,
|
|
"num_tokens": 13507644.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 6.111276054382325,
|
|
"epoch": 6.85927399303829,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017170553721141253,
|
|
"loss": 5.2531,
|
|
"mean_token_accuracy": 0.2076312780380249,
|
|
"num_tokens": 13516713.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 6.111093664169312,
|
|
"epoch": 6.864246643460965,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001713587498857289,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.21678782850503922,
|
|
"num_tokens": 13525603.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 6.1384512424469,
|
|
"epoch": 6.86921929388364,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017101227479040498,
|
|
"loss": 5.229,
|
|
"mean_token_accuracy": 0.20703891068696975,
|
|
"num_tokens": 13534809.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 6.078368043899536,
|
|
"epoch": 6.8741919443063155,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017066611296923424,
|
|
"loss": 5.2037,
|
|
"mean_token_accuracy": 0.21200344413518907,
|
|
"num_tokens": 13544482.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 6.0951861381530765,
|
|
"epoch": 6.879164594728991,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017032026546506595,
|
|
"loss": 5.2409,
|
|
"mean_token_accuracy": 0.21056111603975297,
|
|
"num_tokens": 13553583.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 6.1421895027160645,
|
|
"epoch": 6.884137245151666,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016997473331980295,
|
|
"loss": 5.3261,
|
|
"mean_token_accuracy": 0.20339266508817672,
|
|
"num_tokens": 13562387.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 6.1881529808044435,
|
|
"epoch": 6.889109895574341,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001696295175743977,
|
|
"loss": 5.3418,
|
|
"mean_token_accuracy": 0.20739551186561583,
|
|
"num_tokens": 13572654.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 6.145324945449829,
|
|
"epoch": 6.8940825459970165,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016928461926884954,
|
|
"loss": 5.2131,
|
|
"mean_token_accuracy": 0.21486938148736953,
|
|
"num_tokens": 13582066.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 6.098127126693726,
|
|
"epoch": 6.899055196419692,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00016894003944220167,
|
|
"loss": 5.2014,
|
|
"mean_token_accuracy": 0.21641439497470855,
|
|
"num_tokens": 13591702.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 6.116765403747559,
|
|
"epoch": 6.904027846842367,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001685957791325375,
|
|
"loss": 5.2497,
|
|
"mean_token_accuracy": 0.2123704209923744,
|
|
"num_tokens": 13601822.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.935379505157471,
|
|
"epoch": 6.909000497265042,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001682518393769782,
|
|
"loss": 5.0261,
|
|
"mean_token_accuracy": 0.23001741170883178,
|
|
"num_tokens": 13611389.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 6.024700880050659,
|
|
"epoch": 6.9139731476877175,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016790822121167913,
|
|
"loss": 5.2189,
|
|
"mean_token_accuracy": 0.21858084946870804,
|
|
"num_tokens": 13621590.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 6.035378885269165,
|
|
"epoch": 6.918945798110393,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016756492567182674,
|
|
"loss": 5.1457,
|
|
"mean_token_accuracy": 0.21565153002738952,
|
|
"num_tokens": 13630319.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 6.067088842391968,
|
|
"epoch": 6.923918448533068,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00016722195379163568,
|
|
"loss": 5.1391,
|
|
"mean_token_accuracy": 0.22164399921894073,
|
|
"num_tokens": 13639807.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 6.0557966232299805,
|
|
"epoch": 6.928891098955743,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00016687930660434528,
|
|
"loss": 5.171,
|
|
"mean_token_accuracy": 0.2240081772208214,
|
|
"num_tokens": 13649522.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 6.07701268196106,
|
|
"epoch": 6.9338637493784185,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00016653698514221694,
|
|
"loss": 5.1751,
|
|
"mean_token_accuracy": 0.2173178970813751,
|
|
"num_tokens": 13658855.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.9846930503845215,
|
|
"epoch": 6.938836399801094,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001661949904365308,
|
|
"loss": 5.0729,
|
|
"mean_token_accuracy": 0.23004183173179626,
|
|
"num_tokens": 13669196.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 6.090537643432617,
|
|
"epoch": 6.943809050223769,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001658533235175824,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.21750414073467256,
|
|
"num_tokens": 13680222.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 6.1159144878387455,
|
|
"epoch": 6.948781700646444,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016551198541468,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.21480392515659333,
|
|
"num_tokens": 13689392.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 6.062662744522095,
|
|
"epoch": 6.9537543510691195,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016517097715614115,
|
|
"loss": 5.1226,
|
|
"mean_token_accuracy": 0.22377638071775435,
|
|
"num_tokens": 13698104.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 6.116061639785767,
|
|
"epoch": 6.958727001491795,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016483029976928958,
|
|
"loss": 5.2571,
|
|
"mean_token_accuracy": 0.21731143146753312,
|
|
"num_tokens": 13707386.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.958727001491795,
|
|
"eval_entropy": 5.9695291420281755,
|
|
"eval_loss": 6.435119152069092,
|
|
"eval_mean_token_accuracy": 0.15086227360992663,
|
|
"eval_num_tokens": 13707386.0,
|
|
"eval_runtime": 1.926,
|
|
"eval_samples_per_second": 1799.089,
|
|
"eval_steps_per_second": 225.34,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 6.094450664520264,
|
|
"epoch": 6.96369965191447,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001644899542804526,
|
|
"loss": 5.1723,
|
|
"mean_token_accuracy": 0.21773437261581421,
|
|
"num_tokens": 13716643.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.910398817062378,
|
|
"epoch": 6.968672302337145,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016414994171495728,
|
|
"loss": 5.0874,
|
|
"mean_token_accuracy": 0.23121197074651717,
|
|
"num_tokens": 13727125.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 6.084078359603882,
|
|
"epoch": 6.973644952759821,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016381026309712787,
|
|
"loss": 5.2737,
|
|
"mean_token_accuracy": 0.21279190927743913,
|
|
"num_tokens": 13737267.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 6.137829113006592,
|
|
"epoch": 6.978617603182496,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016347091945028273,
|
|
"loss": 5.3014,
|
|
"mean_token_accuracy": 0.20867679715156556,
|
|
"num_tokens": 13748051.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 6.136505651473999,
|
|
"epoch": 6.983590253605172,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001631319117967307,
|
|
"loss": 5.2523,
|
|
"mean_token_accuracy": 0.20952174812555313,
|
|
"num_tokens": 13758958.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 6.033654737472534,
|
|
"epoch": 6.988562904027847,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016279324115776886,
|
|
"loss": 5.0876,
|
|
"mean_token_accuracy": 0.23830010145902633,
|
|
"num_tokens": 13768566.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 6.083563566207886,
|
|
"epoch": 6.993535554450522,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001624549085536785,
|
|
"loss": 5.2883,
|
|
"mean_token_accuracy": 0.2105501502752304,
|
|
"num_tokens": 13777777.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 6.015117835998535,
|
|
"epoch": 6.998508204873198,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016211691500372304,
|
|
"loss": 5.248,
|
|
"mean_token_accuracy": 0.22078390121459962,
|
|
"num_tokens": 13788178.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.934685283237034,
|
|
"epoch": 7.0029835902536055,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016177926152614414,
|
|
"loss": 4.9597,
|
|
"mean_token_accuracy": 0.24510053131315443,
|
|
"num_tokens": 13797143.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.9320894241333,
|
|
"epoch": 7.007956240676281,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016144194913815887,
|
|
"loss": 4.8863,
|
|
"mean_token_accuracy": 0.2524819880723953,
|
|
"num_tokens": 13806835.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 6.125775480270386,
|
|
"epoch": 7.012928891098956,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001611049788559572,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.22054606080055236,
|
|
"num_tokens": 13816313.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 6.108788537979126,
|
|
"epoch": 7.017901541521631,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016076835169469796,
|
|
"loss": 5.1372,
|
|
"mean_token_accuracy": 0.2231244057416916,
|
|
"num_tokens": 13825692.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 6.148027086257935,
|
|
"epoch": 7.0228741919443065,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00016043206866850652,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.21890397369861603,
|
|
"num_tokens": 13835398.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 6.044643926620483,
|
|
"epoch": 7.027846842366982,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001600961307904717,
|
|
"loss": 5.0488,
|
|
"mean_token_accuracy": 0.22903948724269868,
|
|
"num_tokens": 13844548.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 6.104455757141113,
|
|
"epoch": 7.032819492789657,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015976053907264208,
|
|
"loss": 5.0942,
|
|
"mean_token_accuracy": 0.22442494034767152,
|
|
"num_tokens": 13853844.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 6.064545011520385,
|
|
"epoch": 7.037792143212332,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015942529452602378,
|
|
"loss": 5.0299,
|
|
"mean_token_accuracy": 0.23745042085647583,
|
|
"num_tokens": 13862477.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 6.063648128509522,
|
|
"epoch": 7.0427647936350075,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00015909039816057674,
|
|
"loss": 4.9885,
|
|
"mean_token_accuracy": 0.24085533767938613,
|
|
"num_tokens": 13874248.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.964374780654907,
|
|
"epoch": 7.047737444057683,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015875585098521218,
|
|
"loss": 4.8877,
|
|
"mean_token_accuracy": 0.2512317210435867,
|
|
"num_tokens": 13883952.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.988623476028442,
|
|
"epoch": 7.052710094480358,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015842165400778942,
|
|
"loss": 4.9548,
|
|
"mean_token_accuracy": 0.24423868209123611,
|
|
"num_tokens": 13893782.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.907321882247925,
|
|
"epoch": 7.057682744903033,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001580878082351123,
|
|
"loss": 4.9007,
|
|
"mean_token_accuracy": 0.25012180507183074,
|
|
"num_tokens": 13903366.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 6.070351457595825,
|
|
"epoch": 7.0626553953257085,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015775431467292733,
|
|
"loss": 5.1129,
|
|
"mean_token_accuracy": 0.2305818408727646,
|
|
"num_tokens": 13912345.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 6.115953159332276,
|
|
"epoch": 7.067628045748384,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00015742117432591935,
|
|
"loss": 5.1103,
|
|
"mean_token_accuracy": 0.22188520580530166,
|
|
"num_tokens": 13922670.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.988445234298706,
|
|
"epoch": 7.072600696171059,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00015708838819770922,
|
|
"loss": 4.9847,
|
|
"mean_token_accuracy": 0.2422062039375305,
|
|
"num_tokens": 13932367.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 6.001406621932984,
|
|
"epoch": 7.077573346593734,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000156755957290851,
|
|
"loss": 5.0123,
|
|
"mean_token_accuracy": 0.24380915462970734,
|
|
"num_tokens": 13942996.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 6.187489938735962,
|
|
"epoch": 7.0825459970164095,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015642388260682816,
|
|
"loss": 5.122,
|
|
"mean_token_accuracy": 0.22447860687971116,
|
|
"num_tokens": 13952296.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 6.116701412200928,
|
|
"epoch": 7.087518647439085,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015609216514605136,
|
|
"loss": 5.1407,
|
|
"mean_token_accuracy": 0.22354583889245988,
|
|
"num_tokens": 13962169.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 6.091638231277466,
|
|
"epoch": 7.09249129786176,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015576080590785503,
|
|
"loss": 5.1445,
|
|
"mean_token_accuracy": 0.22529662102460862,
|
|
"num_tokens": 13971184.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 6.1126546382904055,
|
|
"epoch": 7.097463948284435,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015542980589049413,
|
|
"loss": 5.1986,
|
|
"mean_token_accuracy": 0.22777486890554427,
|
|
"num_tokens": 13980379.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 6.040165948867798,
|
|
"epoch": 7.1024365987071105,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00015509916609114188,
|
|
"loss": 5.0566,
|
|
"mean_token_accuracy": 0.2311932161450386,
|
|
"num_tokens": 13989405.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 6.148871231079101,
|
|
"epoch": 7.107409249129786,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00015476888750588585,
|
|
"loss": 5.1632,
|
|
"mean_token_accuracy": 0.21919876784086229,
|
|
"num_tokens": 13998208.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 6.176788854598999,
|
|
"epoch": 7.112381899552461,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015443897112972572,
|
|
"loss": 5.2107,
|
|
"mean_token_accuracy": 0.2186177432537079,
|
|
"num_tokens": 14007041.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 6.112042617797852,
|
|
"epoch": 7.117354549975137,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015410941795657008,
|
|
"loss": 5.0685,
|
|
"mean_token_accuracy": 0.23355651944875716,
|
|
"num_tokens": 14017637.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 6.116136646270752,
|
|
"epoch": 7.122327200397812,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00015378022897923277,
|
|
"loss": 5.2266,
|
|
"mean_token_accuracy": 0.212307907640934,
|
|
"num_tokens": 14027532.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 6.122160959243774,
|
|
"epoch": 7.127299850820488,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00015345140518943103,
|
|
"loss": 5.1934,
|
|
"mean_token_accuracy": 0.21604256927967072,
|
|
"num_tokens": 14037442.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 6.164898824691773,
|
|
"epoch": 7.132272501243163,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001531229475777818,
|
|
"loss": 5.1798,
|
|
"mean_token_accuracy": 0.21570329368114471,
|
|
"num_tokens": 14048898.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 6.097230243682861,
|
|
"epoch": 7.137245151665838,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001527948571337986,
|
|
"loss": 5.152,
|
|
"mean_token_accuracy": 0.21677271276712418,
|
|
"num_tokens": 14058594.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.9200780391693115,
|
|
"epoch": 7.142217802088513,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001524671348458892,
|
|
"loss": 4.8931,
|
|
"mean_token_accuracy": 0.24753040969371795,
|
|
"num_tokens": 14068254.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 6.064241218566894,
|
|
"epoch": 7.147190452511189,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00015213978170135187,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.22700260430574418,
|
|
"num_tokens": 14077158.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 6.104225969314575,
|
|
"epoch": 7.152163102933864,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015181279868637312,
|
|
"loss": 5.1289,
|
|
"mean_token_accuracy": 0.22088966220617295,
|
|
"num_tokens": 14086737.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 6.165177488327027,
|
|
"epoch": 7.157135753356539,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015148618678602433,
|
|
"loss": 5.239,
|
|
"mean_token_accuracy": 0.2140037700533867,
|
|
"num_tokens": 14096606.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 6.028970098495483,
|
|
"epoch": 7.162108403779214,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00015115994698425872,
|
|
"loss": 5.1098,
|
|
"mean_token_accuracy": 0.22850220054388046,
|
|
"num_tokens": 14107125.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 6.083437490463257,
|
|
"epoch": 7.16708105420189,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001508340802639087,
|
|
"loss": 5.1886,
|
|
"mean_token_accuracy": 0.21969438344240189,
|
|
"num_tokens": 14117665.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 6.028297185897827,
|
|
"epoch": 7.172053704624565,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00015050858760668246,
|
|
"loss": 4.9995,
|
|
"mean_token_accuracy": 0.23807610124349593,
|
|
"num_tokens": 14128730.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 6.074712896347046,
|
|
"epoch": 7.17702635504724,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015018346999316173,
|
|
"loss": 5.1732,
|
|
"mean_token_accuracy": 0.22429172098636627,
|
|
"num_tokens": 14139670.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 6.108974599838257,
|
|
"epoch": 7.181999005469915,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000149858728402798,
|
|
"loss": 5.1413,
|
|
"mean_token_accuracy": 0.21930166482925414,
|
|
"num_tokens": 14149308.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 6.066647958755493,
|
|
"epoch": 7.186971655892591,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00014953436381390998,
|
|
"loss": 5.0529,
|
|
"mean_token_accuracy": 0.2287716880440712,
|
|
"num_tokens": 14160280.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 6.160626697540283,
|
|
"epoch": 7.191944306315266,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014921037720368085,
|
|
"loss": 5.2253,
|
|
"mean_token_accuracy": 0.2128698229789734,
|
|
"num_tokens": 14170313.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 6.088065671920776,
|
|
"epoch": 7.196916956737941,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014888676954815484,
|
|
"loss": 5.1151,
|
|
"mean_token_accuracy": 0.2273606464266777,
|
|
"num_tokens": 14180418.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 6.075307083129883,
|
|
"epoch": 7.201889607160616,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001485635418222347,
|
|
"loss": 5.1366,
|
|
"mean_token_accuracy": 0.22155166715383529,
|
|
"num_tokens": 14190503.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 6.1594908237457275,
|
|
"epoch": 7.206862257583292,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014824069499967863,
|
|
"loss": 5.1276,
|
|
"mean_token_accuracy": 0.2196408897638321,
|
|
"num_tokens": 14200193.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.904731130599975,
|
|
"epoch": 7.211834908005967,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001479182300530971,
|
|
"loss": 4.898,
|
|
"mean_token_accuracy": 0.24749952852725982,
|
|
"num_tokens": 14209601.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 6.0823732852935795,
|
|
"epoch": 7.216807558428642,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014759614795395044,
|
|
"loss": 5.0388,
|
|
"mean_token_accuracy": 0.2321030765771866,
|
|
"num_tokens": 14218966.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 6.066477727890015,
|
|
"epoch": 7.221780208851317,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00014727444967254527,
|
|
"loss": 5.0962,
|
|
"mean_token_accuracy": 0.22423945516347885,
|
|
"num_tokens": 14228805.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 6.078905391693115,
|
|
"epoch": 7.226752859273993,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001469531361780322,
|
|
"loss": 5.0896,
|
|
"mean_token_accuracy": 0.2307385101914406,
|
|
"num_tokens": 14237719.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 6.079478740692139,
|
|
"epoch": 7.231725509696668,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014663220843840266,
|
|
"loss": 5.0392,
|
|
"mean_token_accuracy": 0.23432991653680801,
|
|
"num_tokens": 14247496.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.957202625274658,
|
|
"epoch": 7.236698160119343,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001463116674204855,
|
|
"loss": 4.9181,
|
|
"mean_token_accuracy": 0.24902428090572357,
|
|
"num_tokens": 14257715.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 6.074948120117187,
|
|
"epoch": 7.241670810542018,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014599151408994503,
|
|
"loss": 5.1543,
|
|
"mean_token_accuracy": 0.2269188269972801,
|
|
"num_tokens": 14267546.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 6.0178498268127445,
|
|
"epoch": 7.246643460964695,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014567174941127752,
|
|
"loss": 5.0418,
|
|
"mean_token_accuracy": 0.23460027128458022,
|
|
"num_tokens": 14276808.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 6.082131195068359,
|
|
"epoch": 7.25161611138737,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00014535237434780814,
|
|
"loss": 5.1574,
|
|
"mean_token_accuracy": 0.22340013235807418,
|
|
"num_tokens": 14286467.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 6.184965419769287,
|
|
"epoch": 7.256588761810045,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001450333898616885,
|
|
"loss": 5.254,
|
|
"mean_token_accuracy": 0.21139246970415115,
|
|
"num_tokens": 14296308.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 6.004900598526001,
|
|
"epoch": 7.26156141223272,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014471479691389343,
|
|
"loss": 4.9865,
|
|
"mean_token_accuracy": 0.23349099308252336,
|
|
"num_tokens": 14307267.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 6.101950979232788,
|
|
"epoch": 7.266534062655396,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014439659646421843,
|
|
"loss": 5.1624,
|
|
"mean_token_accuracy": 0.21974574625492097,
|
|
"num_tokens": 14317053.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 6.060114336013794,
|
|
"epoch": 7.271506713078071,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001440787894712764,
|
|
"loss": 5.1097,
|
|
"mean_token_accuracy": 0.22517394572496413,
|
|
"num_tokens": 14327123.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 6.12006778717041,
|
|
"epoch": 7.276479363500746,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001437613768924948,
|
|
"loss": 5.1403,
|
|
"mean_token_accuracy": 0.22170850187540053,
|
|
"num_tokens": 14337552.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.885631847381592,
|
|
"epoch": 7.281452013923421,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014344435968411317,
|
|
"loss": 4.9211,
|
|
"mean_token_accuracy": 0.24753690809011458,
|
|
"num_tokens": 14347692.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 6.114700222015381,
|
|
"epoch": 7.286424664346097,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001431277388011796,
|
|
"loss": 5.155,
|
|
"mean_token_accuracy": 0.22157864272594452,
|
|
"num_tokens": 14357306.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 6.159466505050659,
|
|
"epoch": 7.291397314768772,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00014281151519754853,
|
|
"loss": 5.1531,
|
|
"mean_token_accuracy": 0.21165740489959717,
|
|
"num_tokens": 14367914.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 6.179617166519165,
|
|
"epoch": 7.296369965191447,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001424956898258774,
|
|
"loss": 5.2625,
|
|
"mean_token_accuracy": 0.21376026868820192,
|
|
"num_tokens": 14377852.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 6.100818300247193,
|
|
"epoch": 7.301342615614122,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00014218026363762383,
|
|
"loss": 5.1658,
|
|
"mean_token_accuracy": 0.2205379232764244,
|
|
"num_tokens": 14388087.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 6.094323921203613,
|
|
"epoch": 7.306315266036798,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001418652375830431,
|
|
"loss": 5.0924,
|
|
"mean_token_accuracy": 0.2314886197447777,
|
|
"num_tokens": 14399060.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 6.059122753143311,
|
|
"epoch": 7.311287916459473,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014155061261118487,
|
|
"loss": 5.0878,
|
|
"mean_token_accuracy": 0.22482530921697616,
|
|
"num_tokens": 14408975.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 6.058581972122193,
|
|
"epoch": 7.316260566882148,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014123638966989055,
|
|
"loss": 5.188,
|
|
"mean_token_accuracy": 0.21276942938566207,
|
|
"num_tokens": 14418932.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 6.06401572227478,
|
|
"epoch": 7.321233217304823,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001409225697057904,
|
|
"loss": 5.1023,
|
|
"mean_token_accuracy": 0.21949893981218338,
|
|
"num_tokens": 14428441.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 6.02791953086853,
|
|
"epoch": 7.326205867727499,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001406091536643006,
|
|
"loss": 5.0464,
|
|
"mean_token_accuracy": 0.23075195103883744,
|
|
"num_tokens": 14438849.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 6.0650403022766115,
|
|
"epoch": 7.331178518150174,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001402961424896206,
|
|
"loss": 5.1725,
|
|
"mean_token_accuracy": 0.21486416161060334,
|
|
"num_tokens": 14448927.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 6.114551687240601,
|
|
"epoch": 7.336151168572849,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013998353712473005,
|
|
"loss": 5.175,
|
|
"mean_token_accuracy": 0.22162754833698273,
|
|
"num_tokens": 14458754.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 6.078936147689819,
|
|
"epoch": 7.341123818995524,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013967133851138603,
|
|
"loss": 5.2,
|
|
"mean_token_accuracy": 0.21388146430253982,
|
|
"num_tokens": 14469752.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 6.132298326492309,
|
|
"epoch": 7.3460964694182,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00013935954759012036,
|
|
"loss": 5.1605,
|
|
"mean_token_accuracy": 0.21429084688425065,
|
|
"num_tokens": 14479672.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 6.062038612365723,
|
|
"epoch": 7.351069119840875,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001390481653002365,
|
|
"loss": 5.0419,
|
|
"mean_token_accuracy": 0.23187634348869324,
|
|
"num_tokens": 14490428.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 6.091402864456176,
|
|
"epoch": 7.35604177026355,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001387371925798069,
|
|
"loss": 5.0993,
|
|
"mean_token_accuracy": 0.22358826100826262,
|
|
"num_tokens": 14499736.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 6.061456489562988,
|
|
"epoch": 7.361014420686225,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00013842663036567026,
|
|
"loss": 5.11,
|
|
"mean_token_accuracy": 0.22887204736471176,
|
|
"num_tokens": 14509420.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 6.046993589401245,
|
|
"epoch": 7.365987071108901,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013811647959342834,
|
|
"loss": 5.1042,
|
|
"mean_token_accuracy": 0.23050911873579025,
|
|
"num_tokens": 14519994.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 6.039501428604126,
|
|
"epoch": 7.370959721531577,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013780674119744375,
|
|
"loss": 5.0751,
|
|
"mean_token_accuracy": 0.22759976834058762,
|
|
"num_tokens": 14528283.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 6.063188791275024,
|
|
"epoch": 7.375932371954252,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001374974161108364,
|
|
"loss": 5.1073,
|
|
"mean_token_accuracy": 0.22587334513664245,
|
|
"num_tokens": 14537612.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 6.067985200881958,
|
|
"epoch": 7.380905022376927,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001371885052654812,
|
|
"loss": 5.1261,
|
|
"mean_token_accuracy": 0.22407792806625365,
|
|
"num_tokens": 14547820.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 6.072678661346435,
|
|
"epoch": 7.3858776727996025,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001368800095920053,
|
|
"loss": 5.0871,
|
|
"mean_token_accuracy": 0.23270974159240723,
|
|
"num_tokens": 14557123.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 6.088122081756592,
|
|
"epoch": 7.390850323222278,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001365719300197848,
|
|
"loss": 5.115,
|
|
"mean_token_accuracy": 0.2217855453491211,
|
|
"num_tokens": 14565736.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 6.065373706817627,
|
|
"epoch": 7.395822973644953,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013626426747694254,
|
|
"loss": 5.0912,
|
|
"mean_token_accuracy": 0.21954645216464996,
|
|
"num_tokens": 14575454.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 6.035569190979004,
|
|
"epoch": 7.400795624067628,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013595702289034473,
|
|
"loss": 5.0305,
|
|
"mean_token_accuracy": 0.22974644750356674,
|
|
"num_tokens": 14585041.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 6.088772630691528,
|
|
"epoch": 7.4057682744903035,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013565019718559854,
|
|
"loss": 5.0549,
|
|
"mean_token_accuracy": 0.2285393461585045,
|
|
"num_tokens": 14594829.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 6.108497285842896,
|
|
"epoch": 7.410740924912979,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00013534379128704937,
|
|
"loss": 5.2053,
|
|
"mean_token_accuracy": 0.21552834510803223,
|
|
"num_tokens": 14605271.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 6.034263610839844,
|
|
"epoch": 7.415713575335654,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013503780611777766,
|
|
"loss": 5.0372,
|
|
"mean_token_accuracy": 0.22839145213365555,
|
|
"num_tokens": 14616002.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 6.17641134262085,
|
|
"epoch": 7.420686225758329,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001347322425995966,
|
|
"loss": 5.2505,
|
|
"mean_token_accuracy": 0.21284613460302354,
|
|
"num_tokens": 14626042.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 6.128395509719849,
|
|
"epoch": 7.4256588761810045,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013442710165304894,
|
|
"loss": 5.1635,
|
|
"mean_token_accuracy": 0.22012544870376588,
|
|
"num_tokens": 14636367.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 6.147053527832031,
|
|
"epoch": 7.43063152660368,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00013412238419740444,
|
|
"loss": 5.2007,
|
|
"mean_token_accuracy": 0.21472179591655732,
|
|
"num_tokens": 14646596.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 6.077463006973266,
|
|
"epoch": 7.435604177026355,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013381809115065716,
|
|
"loss": 5.1562,
|
|
"mean_token_accuracy": 0.22098080813884735,
|
|
"num_tokens": 14656757.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 6.149050235748291,
|
|
"epoch": 7.44057682744903,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013351422342952228,
|
|
"loss": 5.1737,
|
|
"mean_token_accuracy": 0.20958020091056823,
|
|
"num_tokens": 14666060.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 6.081441164016724,
|
|
"epoch": 7.4455494778717055,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013321078194943392,
|
|
"loss": 5.1284,
|
|
"mean_token_accuracy": 0.22348977327346803,
|
|
"num_tokens": 14675677.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 6.024363279342651,
|
|
"epoch": 7.450522128294381,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001329077676245421,
|
|
"loss": 5.0428,
|
|
"mean_token_accuracy": 0.23450805693864823,
|
|
"num_tokens": 14684585.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 6.076003837585449,
|
|
"epoch": 7.455494778717056,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00013260518136770983,
|
|
"loss": 5.2137,
|
|
"mean_token_accuracy": 0.2148275136947632,
|
|
"num_tokens": 14694853.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 6.119724988937378,
|
|
"epoch": 7.460467429139731,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013230302409051066,
|
|
"loss": 5.2136,
|
|
"mean_token_accuracy": 0.21798385828733444,
|
|
"num_tokens": 14704600.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 6.093146705627442,
|
|
"epoch": 7.4654400795624065,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013200129670322563,
|
|
"loss": 5.1252,
|
|
"mean_token_accuracy": 0.23004495352506638,
|
|
"num_tokens": 14714476.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 6.085436773300171,
|
|
"epoch": 7.470412729985082,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013170000011484084,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.2207262247800827,
|
|
"num_tokens": 14723704.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 6.138432931900025,
|
|
"epoch": 7.475385380407757,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013139913523304458,
|
|
"loss": 5.2103,
|
|
"mean_token_accuracy": 0.21499937921762466,
|
|
"num_tokens": 14734128.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 6.09359335899353,
|
|
"epoch": 7.480358030830432,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013109870296422436,
|
|
"loss": 5.1828,
|
|
"mean_token_accuracy": 0.21726346164941787,
|
|
"num_tokens": 14743186.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.9387526512146,
|
|
"epoch": 7.4853306812531075,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00013079870421346465,
|
|
"loss": 4.9251,
|
|
"mean_token_accuracy": 0.25106638073921206,
|
|
"num_tokens": 14752315.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 6.078979253768921,
|
|
"epoch": 7.490303331675783,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013049913988454385,
|
|
"loss": 5.1153,
|
|
"mean_token_accuracy": 0.2258251950144768,
|
|
"num_tokens": 14761006.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 6.09332971572876,
|
|
"epoch": 7.495275982098459,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013020001087993142,
|
|
"loss": 5.096,
|
|
"mean_token_accuracy": 0.22136687934398652,
|
|
"num_tokens": 14770685.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 6.0706840515136715,
|
|
"epoch": 7.500248632521133,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00012990131810078565,
|
|
"loss": 5.1399,
|
|
"mean_token_accuracy": 0.22274822741746902,
|
|
"num_tokens": 14781316.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 6.089400577545166,
|
|
"epoch": 7.505221282943809,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00012960306244695041,
|
|
"loss": 5.0844,
|
|
"mean_token_accuracy": 0.22578289806842805,
|
|
"num_tokens": 14789877.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.938261795043945,
|
|
"epoch": 7.510193933366485,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00012930524481695278,
|
|
"loss": 4.9756,
|
|
"mean_token_accuracy": 0.23895589262247086,
|
|
"num_tokens": 14800901.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 6.067863273620605,
|
|
"epoch": 7.51516658378916,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012900786610800047,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.2315218433737755,
|
|
"num_tokens": 14811055.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 6.043261957168579,
|
|
"epoch": 7.520139234211835,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00012871092721597832,
|
|
"loss": 5.1029,
|
|
"mean_token_accuracy": 0.2150714084506035,
|
|
"num_tokens": 14820305.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.921695804595947,
|
|
"epoch": 7.52511188463451,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012841442903544666,
|
|
"loss": 5.0301,
|
|
"mean_token_accuracy": 0.2393841862678528,
|
|
"num_tokens": 14829746.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 6.060308980941772,
|
|
"epoch": 7.530084535057186,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00012811837245963817,
|
|
"loss": 5.1581,
|
|
"mean_token_accuracy": 0.22447711378335952,
|
|
"num_tokens": 14837880.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 6.108771419525146,
|
|
"epoch": 7.535057185479861,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012782275838045475,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.21333229690790176,
|
|
"num_tokens": 14848543.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 6.049766921997071,
|
|
"epoch": 7.540029835902536,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00012752758768846557,
|
|
"loss": 5.0326,
|
|
"mean_token_accuracy": 0.2269174799323082,
|
|
"num_tokens": 14858063.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 6.006576061248779,
|
|
"epoch": 7.545002486325211,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012723286127290387,
|
|
"loss": 5.0539,
|
|
"mean_token_accuracy": 0.23105744868516923,
|
|
"num_tokens": 14868166.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 6.109930324554443,
|
|
"epoch": 7.549975136747887,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00012693858002166465,
|
|
"loss": 5.1813,
|
|
"mean_token_accuracy": 0.21830514818429947,
|
|
"num_tokens": 14879315.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 6.142914724349976,
|
|
"epoch": 7.554947787170562,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012664474482130145,
|
|
"loss": 5.1984,
|
|
"mean_token_accuracy": 0.21337404549121858,
|
|
"num_tokens": 14889433.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.975914239883423,
|
|
"epoch": 7.559920437593237,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012635135655702446,
|
|
"loss": 5.0268,
|
|
"mean_token_accuracy": 0.24069226384162903,
|
|
"num_tokens": 14900143.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 6.056736040115356,
|
|
"epoch": 7.564893088015912,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00012605841611269721,
|
|
"loss": 5.0436,
|
|
"mean_token_accuracy": 0.2327452629804611,
|
|
"num_tokens": 14910295.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 6.112710666656494,
|
|
"epoch": 7.569865738438588,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001257659243708341,
|
|
"loss": 5.1014,
|
|
"mean_token_accuracy": 0.2299015626311302,
|
|
"num_tokens": 14919767.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 6.13430233001709,
|
|
"epoch": 7.574838388861263,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001254738822125978,
|
|
"loss": 5.1833,
|
|
"mean_token_accuracy": 0.2198774039745331,
|
|
"num_tokens": 14929925.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 6.104534769058228,
|
|
"epoch": 7.579811039283938,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012518229051779666,
|
|
"loss": 5.1317,
|
|
"mean_token_accuracy": 0.22698255032300949,
|
|
"num_tokens": 14938765.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 6.068631315231324,
|
|
"epoch": 7.584783689706613,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001248911501648817,
|
|
"loss": 5.1075,
|
|
"mean_token_accuracy": 0.2205242320895195,
|
|
"num_tokens": 14948926.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 6.079456520080567,
|
|
"epoch": 7.589756340129289,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012460046203094458,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.22073392421007157,
|
|
"num_tokens": 14958841.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 6.0667585849761965,
|
|
"epoch": 7.594728990551964,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001243102269917143,
|
|
"loss": 5.1539,
|
|
"mean_token_accuracy": 0.21743342131376267,
|
|
"num_tokens": 14967973.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 6.09203462600708,
|
|
"epoch": 7.599701640974639,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00012402044592155497,
|
|
"loss": 5.1008,
|
|
"mean_token_accuracy": 0.22466561794281006,
|
|
"num_tokens": 14977877.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 6.12176365852356,
|
|
"epoch": 7.604674291397314,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00012373111969346323,
|
|
"loss": 5.1396,
|
|
"mean_token_accuracy": 0.22089193910360336,
|
|
"num_tokens": 14987943.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 6.145666885375976,
|
|
"epoch": 7.60964694181999,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00012344224917906516,
|
|
"loss": 5.1864,
|
|
"mean_token_accuracy": 0.21757356971502304,
|
|
"num_tokens": 14998685.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 6.1193612575531,
|
|
"epoch": 7.614619592242665,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001231538352486142,
|
|
"loss": 5.1398,
|
|
"mean_token_accuracy": 0.23023944199085236,
|
|
"num_tokens": 15008258.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 6.090306901931763,
|
|
"epoch": 7.619592242665341,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012286587877098808,
|
|
"loss": 5.1572,
|
|
"mean_token_accuracy": 0.21699107736349105,
|
|
"num_tokens": 15017738.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 6.089849901199341,
|
|
"epoch": 7.624564893088015,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012257838061368657,
|
|
"loss": 5.1495,
|
|
"mean_token_accuracy": 0.2186768725514412,
|
|
"num_tokens": 15026834.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.9421693801879885,
|
|
"epoch": 7.629537543510692,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012229134164282884,
|
|
"loss": 4.9719,
|
|
"mean_token_accuracy": 0.24253862649202346,
|
|
"num_tokens": 15036811.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 6.09283676147461,
|
|
"epoch": 7.634510193933367,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00012200476272315015,
|
|
"loss": 5.1445,
|
|
"mean_token_accuracy": 0.22076599895954133,
|
|
"num_tokens": 15047540.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 6.0262433052062985,
|
|
"epoch": 7.639482844356042,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012171864471800033,
|
|
"loss": 5.0846,
|
|
"mean_token_accuracy": 0.22127593904733658,
|
|
"num_tokens": 15056346.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 6.157078218460083,
|
|
"epoch": 7.644455494778717,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00012143298848934063,
|
|
"loss": 5.2514,
|
|
"mean_token_accuracy": 0.21272394210100173,
|
|
"num_tokens": 15067247.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 6.074346733093262,
|
|
"epoch": 7.649428145201393,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001211477948977408,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.22370686382055283,
|
|
"num_tokens": 15077468.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 6.061591148376465,
|
|
"epoch": 7.654400795624068,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012086306480237723,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.2273955076932907,
|
|
"num_tokens": 15086420.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 6.0737285137176515,
|
|
"epoch": 7.659373446046743,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012057879906102967,
|
|
"loss": 5.1561,
|
|
"mean_token_accuracy": 0.22379403859376906,
|
|
"num_tokens": 15096380.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 6.0712179183959964,
|
|
"epoch": 7.664346096469418,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00012029499853007916,
|
|
"loss": 5.1784,
|
|
"mean_token_accuracy": 0.22324783951044083,
|
|
"num_tokens": 15107362.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 6.138212537765503,
|
|
"epoch": 7.669318746892094,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00012001166406450528,
|
|
"loss": 5.1639,
|
|
"mean_token_accuracy": 0.21822563111782073,
|
|
"num_tokens": 15117749.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 6.1311627388000485,
|
|
"epoch": 7.674291397314769,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00011972879651788323,
|
|
"loss": 5.1411,
|
|
"mean_token_accuracy": 0.22381692379713058,
|
|
"num_tokens": 15128049.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 6.085133361816406,
|
|
"epoch": 7.679264047737444,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011944639674238197,
|
|
"loss": 5.181,
|
|
"mean_token_accuracy": 0.21609336286783218,
|
|
"num_tokens": 15137998.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 6.0769134044647215,
|
|
"epoch": 7.684236698160119,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011916446558876088,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.2244492545723915,
|
|
"num_tokens": 15147380.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 6.065424919128418,
|
|
"epoch": 7.689209348582795,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011888300390636788,
|
|
"loss": 5.0526,
|
|
"mean_token_accuracy": 0.23088915199041365,
|
|
"num_tokens": 15155803.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 6.10335488319397,
|
|
"epoch": 7.69418199900547,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001186020125431364,
|
|
"loss": 5.2135,
|
|
"mean_token_accuracy": 0.2087210774421692,
|
|
"num_tokens": 15166547.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 6.074704599380493,
|
|
"epoch": 7.699154649428145,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00011832149234558286,
|
|
"loss": 5.1043,
|
|
"mean_token_accuracy": 0.2334979310631752,
|
|
"num_tokens": 15176949.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 6.03711838722229,
|
|
"epoch": 7.70412729985082,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00011804144415880445,
|
|
"loss": 5.0299,
|
|
"mean_token_accuracy": 0.23748443871736527,
|
|
"num_tokens": 15186574.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 6.063477611541748,
|
|
"epoch": 7.709099950273496,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011776186882647643,
|
|
"loss": 5.0993,
|
|
"mean_token_accuracy": 0.21924344897270204,
|
|
"num_tokens": 15195626.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 6.042076158523559,
|
|
"epoch": 7.714072600696171,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011748276719084924,
|
|
"loss": 5.1355,
|
|
"mean_token_accuracy": 0.22149912416934966,
|
|
"num_tokens": 15205098.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 6.06373610496521,
|
|
"epoch": 7.719045251118846,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001172041400927466,
|
|
"loss": 5.1035,
|
|
"mean_token_accuracy": 0.22058779746294022,
|
|
"num_tokens": 15214780.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.913583850860595,
|
|
"epoch": 7.724017901541521,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001169259883715623,
|
|
"loss": 4.9817,
|
|
"mean_token_accuracy": 0.24306630790233613,
|
|
"num_tokens": 15224119.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 6.089173746109009,
|
|
"epoch": 7.728990551964197,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001166483128652583,
|
|
"loss": 5.1594,
|
|
"mean_token_accuracy": 0.22122555375099182,
|
|
"num_tokens": 15234795.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 6.13339672088623,
|
|
"epoch": 7.733963202386872,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001163711144103617,
|
|
"loss": 5.1509,
|
|
"mean_token_accuracy": 0.22150615006685256,
|
|
"num_tokens": 15244016.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 6.107003307342529,
|
|
"epoch": 7.738935852809547,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00011609439384196255,
|
|
"loss": 5.1654,
|
|
"mean_token_accuracy": 0.2190254732966423,
|
|
"num_tokens": 15253395.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 6.031355571746826,
|
|
"epoch": 7.743908503232223,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001158181519937113,
|
|
"loss": 5.127,
|
|
"mean_token_accuracy": 0.22414836436510086,
|
|
"num_tokens": 15263184.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 6.08043360710144,
|
|
"epoch": 7.748881153654898,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011554238969781585,
|
|
"loss": 5.0723,
|
|
"mean_token_accuracy": 0.2249154657125473,
|
|
"num_tokens": 15273039.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 6.052745389938354,
|
|
"epoch": 7.753853804077574,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011526710778503977,
|
|
"loss": 5.1186,
|
|
"mean_token_accuracy": 0.21609419584274292,
|
|
"num_tokens": 15281800.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 6.062645959854126,
|
|
"epoch": 7.758826454500249,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011499230708469932,
|
|
"loss": 5.1501,
|
|
"mean_token_accuracy": 0.22063240259885789,
|
|
"num_tokens": 15291249.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 6.104551601409912,
|
|
"epoch": 7.763799104922924,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011471798842466092,
|
|
"loss": 5.147,
|
|
"mean_token_accuracy": 0.22041895389556884,
|
|
"num_tokens": 15301231.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 6.156195259094238,
|
|
"epoch": 7.7687717553455995,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011444415263133905,
|
|
"loss": 5.189,
|
|
"mean_token_accuracy": 0.21967757791280745,
|
|
"num_tokens": 15310947.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.953624534606933,
|
|
"epoch": 7.773744405768275,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011417080052969314,
|
|
"loss": 4.9508,
|
|
"mean_token_accuracy": 0.2327912852168083,
|
|
"num_tokens": 15320376.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.893313217163086,
|
|
"epoch": 7.77871705619095,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011389793294322582,
|
|
"loss": 4.8684,
|
|
"mean_token_accuracy": 0.2520953372120857,
|
|
"num_tokens": 15330268.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 6.050569772720337,
|
|
"epoch": 7.783689706613625,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011362555069397988,
|
|
"loss": 5.0962,
|
|
"mean_token_accuracy": 0.22928949743509291,
|
|
"num_tokens": 15340995.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 6.0523951053619385,
|
|
"epoch": 7.7886623570363005,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011335365460253592,
|
|
"loss": 5.1295,
|
|
"mean_token_accuracy": 0.22712066620588303,
|
|
"num_tokens": 15350799.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.9335620403289795,
|
|
"epoch": 7.793635007458976,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011308224548801014,
|
|
"loss": 4.969,
|
|
"mean_token_accuracy": 0.2395625188946724,
|
|
"num_tokens": 15360472.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 6.121491241455078,
|
|
"epoch": 7.798607657881651,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011281132416805143,
|
|
"loss": 5.2193,
|
|
"mean_token_accuracy": 0.21230860203504562,
|
|
"num_tokens": 15370156.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 6.138769626617432,
|
|
"epoch": 7.803580308304326,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001125408914588394,
|
|
"loss": 5.2628,
|
|
"mean_token_accuracy": 0.20767542272806166,
|
|
"num_tokens": 15380804.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 6.082727241516113,
|
|
"epoch": 7.8085529587270015,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001122709481750814,
|
|
"loss": 5.2043,
|
|
"mean_token_accuracy": 0.21581279188394548,
|
|
"num_tokens": 15390991.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 6.0366603374481205,
|
|
"epoch": 7.813525609149677,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001120014951300105,
|
|
"loss": 5.0677,
|
|
"mean_token_accuracy": 0.22921050935983658,
|
|
"num_tokens": 15400199.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 6.0726637840271,
|
|
"epoch": 7.818498259572352,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00011173253313538284,
|
|
"loss": 5.1115,
|
|
"mean_token_accuracy": 0.22307685017585754,
|
|
"num_tokens": 15410074.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 6.083601808547973,
|
|
"epoch": 7.823470909995027,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001114640630014753,
|
|
"loss": 5.168,
|
|
"mean_token_accuracy": 0.22124277353286742,
|
|
"num_tokens": 15421504.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 6.0955634117126465,
|
|
"epoch": 7.8284435604177025,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011119608553708277,
|
|
"loss": 5.1672,
|
|
"mean_token_accuracy": 0.21694163978099823,
|
|
"num_tokens": 15430469.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 6.027425098419189,
|
|
"epoch": 7.833416210840378,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00011092860154951618,
|
|
"loss": 5.1261,
|
|
"mean_token_accuracy": 0.21827090084552764,
|
|
"num_tokens": 15439690.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 6.13141622543335,
|
|
"epoch": 7.838388861263053,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011066161184459954,
|
|
"loss": 5.235,
|
|
"mean_token_accuracy": 0.21610607355833053,
|
|
"num_tokens": 15448835.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.988946962356567,
|
|
"epoch": 7.843361511685728,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011039511722666795,
|
|
"loss": 4.8779,
|
|
"mean_token_accuracy": 0.2549300894141197,
|
|
"num_tokens": 15459124.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 6.112655973434448,
|
|
"epoch": 7.8483341621084035,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001101291184985651,
|
|
"loss": 5.1283,
|
|
"mean_token_accuracy": 0.22143006920814515,
|
|
"num_tokens": 15469127.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 6.103800201416016,
|
|
"epoch": 7.853306812531079,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010986361646164044,
|
|
"loss": 5.218,
|
|
"mean_token_accuracy": 0.21473403126001359,
|
|
"num_tokens": 15479039.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 6.145141983032227,
|
|
"epoch": 7.858279462953754,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010959861191574746,
|
|
"loss": 5.151,
|
|
"mean_token_accuracy": 0.22080522179603576,
|
|
"num_tokens": 15488186.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 6.140074777603149,
|
|
"epoch": 7.863252113376429,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010933410565924061,
|
|
"loss": 5.1501,
|
|
"mean_token_accuracy": 0.21885878443717957,
|
|
"num_tokens": 15497560.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 6.096857690811158,
|
|
"epoch": 7.868224763799105,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00010907009848897327,
|
|
"loss": 5.168,
|
|
"mean_token_accuracy": 0.22699520438909532,
|
|
"num_tokens": 15507287.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 6.053411149978638,
|
|
"epoch": 7.87319741422178,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010880659120029537,
|
|
"loss": 5.1061,
|
|
"mean_token_accuracy": 0.22361862808465957,
|
|
"num_tokens": 15516642.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 6.065866422653198,
|
|
"epoch": 7.878170064644456,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00010854358458705079,
|
|
"loss": 5.1663,
|
|
"mean_token_accuracy": 0.22471793442964555,
|
|
"num_tokens": 15526426.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 6.06327600479126,
|
|
"epoch": 7.883142715067131,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010828107944157509,
|
|
"loss": 5.1803,
|
|
"mean_token_accuracy": 0.22062380313873292,
|
|
"num_tokens": 15535813.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 6.0963842391967775,
|
|
"epoch": 7.888115365489806,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010801907655469319,
|
|
"loss": 5.1766,
|
|
"mean_token_accuracy": 0.21741364747285843,
|
|
"num_tokens": 15545473.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 6.112465190887451,
|
|
"epoch": 7.893088015912482,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010775757671571675,
|
|
"loss": 5.1823,
|
|
"mean_token_accuracy": 0.22169235348701477,
|
|
"num_tokens": 15556059.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 6.075289011001587,
|
|
"epoch": 7.898060666335157,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010749658071244211,
|
|
"loss": 5.1775,
|
|
"mean_token_accuracy": 0.2181084007024765,
|
|
"num_tokens": 15567252.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 6.109842252731323,
|
|
"epoch": 7.903033316757832,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010723608933114754,
|
|
"loss": 5.119,
|
|
"mean_token_accuracy": 0.23687744438648223,
|
|
"num_tokens": 15577158.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 6.064235258102417,
|
|
"epoch": 7.908005967180507,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010697610335659139,
|
|
"loss": 5.1155,
|
|
"mean_token_accuracy": 0.2271430656313896,
|
|
"num_tokens": 15586153.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 6.08800630569458,
|
|
"epoch": 7.912978617603183,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010671662357200904,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.2192087173461914,
|
|
"num_tokens": 15594732.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 6.084024906158447,
|
|
"epoch": 7.917951268025858,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010645765075911129,
|
|
"loss": 5.0646,
|
|
"mean_token_accuracy": 0.2273509457707405,
|
|
"num_tokens": 15604099.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 6.160137605667114,
|
|
"epoch": 7.922923918448533,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00010619918569808136,
|
|
"loss": 5.2655,
|
|
"mean_token_accuracy": 0.21064961552619935,
|
|
"num_tokens": 15614281.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 6.132921028137207,
|
|
"epoch": 7.927896568871208,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010594122916757288,
|
|
"loss": 5.1945,
|
|
"mean_token_accuracy": 0.21516345143318177,
|
|
"num_tokens": 15624180.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.95137391090393,
|
|
"epoch": 7.932869219293884,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010568378194470754,
|
|
"loss": 4.9342,
|
|
"mean_token_accuracy": 0.23753185868263244,
|
|
"num_tokens": 15633346.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 6.09054160118103,
|
|
"epoch": 7.937841869716559,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00010542684480507278,
|
|
"loss": 5.1044,
|
|
"mean_token_accuracy": 0.2242967888712883,
|
|
"num_tokens": 15643220.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 6.112982559204101,
|
|
"epoch": 7.942814520139234,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010517041852271906,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.2159060776233673,
|
|
"num_tokens": 15653032.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 6.074758672714234,
|
|
"epoch": 7.947787170561909,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00010491450387015813,
|
|
"loss": 5.1595,
|
|
"mean_token_accuracy": 0.2196406364440918,
|
|
"num_tokens": 15662501.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 6.105674695968628,
|
|
"epoch": 7.952759820984585,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010465910161836015,
|
|
"loss": 5.1855,
|
|
"mean_token_accuracy": 0.21278926134109497,
|
|
"num_tokens": 15671387.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 7.952759820984585,
|
|
"eval_entropy": 5.944872314479494,
|
|
"eval_loss": 6.444463729858398,
|
|
"eval_mean_token_accuracy": 0.15110602248717564,
|
|
"eval_num_tokens": 15671387.0,
|
|
"eval_runtime": 1.9286,
|
|
"eval_samples_per_second": 1796.664,
|
|
"eval_steps_per_second": 225.037,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 6.081371879577636,
|
|
"epoch": 7.95773247140726,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010440421253675184,
|
|
"loss": 5.0668,
|
|
"mean_token_accuracy": 0.22586595714092256,
|
|
"num_tokens": 15679939.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 6.06992359161377,
|
|
"epoch": 7.962705121829935,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010414983739321382,
|
|
"loss": 5.1413,
|
|
"mean_token_accuracy": 0.21972243189811708,
|
|
"num_tokens": 15690670.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 6.0315851211547855,
|
|
"epoch": 7.96767777225261,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010389597695407843,
|
|
"loss": 5.0065,
|
|
"mean_token_accuracy": 0.23222112357616426,
|
|
"num_tokens": 15700133.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 6.099147605895996,
|
|
"epoch": 7.972650422675286,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010364263198412746,
|
|
"loss": 5.1041,
|
|
"mean_token_accuracy": 0.22412958294153212,
|
|
"num_tokens": 15709312.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 6.116532659530639,
|
|
"epoch": 7.977623073097961,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010338980324658972,
|
|
"loss": 5.1526,
|
|
"mean_token_accuracy": 0.2124485656619072,
|
|
"num_tokens": 15719122.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 6.155780029296875,
|
|
"epoch": 7.982595723520636,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010313749150313872,
|
|
"loss": 5.215,
|
|
"mean_token_accuracy": 0.21958310753107071,
|
|
"num_tokens": 15729710.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 6.056477451324463,
|
|
"epoch": 7.987568373943311,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00010288569751389082,
|
|
"loss": 5.1197,
|
|
"mean_token_accuracy": 0.2236104726791382,
|
|
"num_tokens": 15738906.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 6.121938467025757,
|
|
"epoch": 7.992541024365988,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010263442203740217,
|
|
"loss": 5.1716,
|
|
"mean_token_accuracy": 0.21789949238300324,
|
|
"num_tokens": 15747647.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 6.082445192337036,
|
|
"epoch": 7.997513674788662,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010238366583066713,
|
|
"loss": 5.1188,
|
|
"mean_token_accuracy": 0.22325633466243744,
|
|
"num_tokens": 15756664.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 6.0865906609429254,
|
|
"epoch": 8.00198906016907,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010213342964911568,
|
|
"loss": 5.0913,
|
|
"mean_token_accuracy": 0.22396356032954323,
|
|
"num_tokens": 15764754.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 6.091531085968017,
|
|
"epoch": 8.006961710591746,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010188371424661095,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.22661810517311096,
|
|
"num_tokens": 15773557.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 6.140142297744751,
|
|
"epoch": 8.01193436101442,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010163452037544745,
|
|
"loss": 5.1764,
|
|
"mean_token_accuracy": 0.2176007643342018,
|
|
"num_tokens": 15782842.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 6.135694360733032,
|
|
"epoch": 8.016907011437096,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001013858487863483,
|
|
"loss": 5.0645,
|
|
"mean_token_accuracy": 0.22746171653270722,
|
|
"num_tokens": 15792334.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 6.1078300952911375,
|
|
"epoch": 8.02187966185977,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001011377002284633,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.2272910326719284,
|
|
"num_tokens": 15800520.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 6.102231502532959,
|
|
"epoch": 8.026852312282447,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010089007544936666,
|
|
"loss": 5.1576,
|
|
"mean_token_accuracy": 0.21926439106464385,
|
|
"num_tokens": 15810399.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.931024646759033,
|
|
"epoch": 8.031824962705121,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00010064297519505435,
|
|
"loss": 4.7818,
|
|
"mean_token_accuracy": 0.2583319410681725,
|
|
"num_tokens": 15820215.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 6.164893388748169,
|
|
"epoch": 8.036797613127797,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010039640020994241,
|
|
"loss": 5.1303,
|
|
"mean_token_accuracy": 0.22190826684236525,
|
|
"num_tokens": 15830187.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.915273141860962,
|
|
"epoch": 8.041770263550472,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00010015035123686443,
|
|
"loss": 4.9325,
|
|
"mean_token_accuracy": 0.24736390858888627,
|
|
"num_tokens": 15840423.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 6.059693956375122,
|
|
"epoch": 8.046742913973148,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.990482901706918e-05,
|
|
"loss": 5.0777,
|
|
"mean_token_accuracy": 0.22998637408018113,
|
|
"num_tokens": 15849700.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 6.100716495513916,
|
|
"epoch": 8.051715564395822,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.965983429021874e-05,
|
|
"loss": 5.1369,
|
|
"mean_token_accuracy": 0.22397322207689285,
|
|
"num_tokens": 15860149.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.956935167312622,
|
|
"epoch": 8.056688214818498,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.941536779438584e-05,
|
|
"loss": 4.9167,
|
|
"mean_token_accuracy": 0.24732605516910552,
|
|
"num_tokens": 15870073.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 6.088764905929565,
|
|
"epoch": 8.061660865241173,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.917143026605201e-05,
|
|
"loss": 5.1112,
|
|
"mean_token_accuracy": 0.22717464566230774,
|
|
"num_tokens": 15881145.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 6.122217607498169,
|
|
"epoch": 8.066633515663849,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.892802244010522e-05,
|
|
"loss": 5.0924,
|
|
"mean_token_accuracy": 0.22754793465137482,
|
|
"num_tokens": 15892588.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 6.083763313293457,
|
|
"epoch": 8.071606166086523,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.86851450498375e-05,
|
|
"loss": 5.0465,
|
|
"mean_token_accuracy": 0.23157695680856705,
|
|
"num_tokens": 15902846.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 6.142479181289673,
|
|
"epoch": 8.0765788165092,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.844279882694309e-05,
|
|
"loss": 5.1198,
|
|
"mean_token_accuracy": 0.22122792154550552,
|
|
"num_tokens": 15912987.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 6.132688331604004,
|
|
"epoch": 8.081551466931876,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.820098450151582e-05,
|
|
"loss": 5.0205,
|
|
"mean_token_accuracy": 0.2357658937573433,
|
|
"num_tokens": 15922808.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 6.088520336151123,
|
|
"epoch": 8.08652411735455,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.795970280204733e-05,
|
|
"loss": 5.0774,
|
|
"mean_token_accuracy": 0.22547124326229095,
|
|
"num_tokens": 15931807.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 6.090492248535156,
|
|
"epoch": 8.091496767777226,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.771895445542453e-05,
|
|
"loss": 5.0766,
|
|
"mean_token_accuracy": 0.2205667808651924,
|
|
"num_tokens": 15941755.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 6.075142240524292,
|
|
"epoch": 8.0964694181999,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.747874018692752e-05,
|
|
"loss": 5.1017,
|
|
"mean_token_accuracy": 0.22342577576637268,
|
|
"num_tokens": 15950112.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 6.086233520507813,
|
|
"epoch": 8.101442068622577,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.723906072022759e-05,
|
|
"loss": 5.0919,
|
|
"mean_token_accuracy": 0.22767400294542312,
|
|
"num_tokens": 15959342.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 6.035185289382935,
|
|
"epoch": 8.106414719045251,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.69999167773847e-05,
|
|
"loss": 5.0052,
|
|
"mean_token_accuracy": 0.2380325675010681,
|
|
"num_tokens": 15968630.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 6.1207301139831545,
|
|
"epoch": 8.111387369467927,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.676130907884564e-05,
|
|
"loss": 5.1612,
|
|
"mean_token_accuracy": 0.22101476192474365,
|
|
"num_tokens": 15978634.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.97645320892334,
|
|
"epoch": 8.116360019890601,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.652323834344166e-05,
|
|
"loss": 5.016,
|
|
"mean_token_accuracy": 0.22978200167417526,
|
|
"num_tokens": 15989450.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 6.095258140563965,
|
|
"epoch": 8.121332670313278,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.628570528838626e-05,
|
|
"loss": 5.0533,
|
|
"mean_token_accuracy": 0.22960386872291566,
|
|
"num_tokens": 15998575.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 6.049364757537842,
|
|
"epoch": 8.126305320735952,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.604871062927328e-05,
|
|
"loss": 5.0373,
|
|
"mean_token_accuracy": 0.2313176989555359,
|
|
"num_tokens": 16007756.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 6.093608713150024,
|
|
"epoch": 8.131277971158628,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.581225508007441e-05,
|
|
"loss": 5.0293,
|
|
"mean_token_accuracy": 0.22788488417863845,
|
|
"num_tokens": 16017274.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 6.073897457122802,
|
|
"epoch": 8.136250621581302,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.557633935313731e-05,
|
|
"loss": 5.1196,
|
|
"mean_token_accuracy": 0.22914549261331557,
|
|
"num_tokens": 16027492.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 6.131371879577637,
|
|
"epoch": 8.141223272003979,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.534096415918354e-05,
|
|
"loss": 5.1068,
|
|
"mean_token_accuracy": 0.22701097130775452,
|
|
"num_tokens": 16036657.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 6.113700103759766,
|
|
"epoch": 8.146195922426653,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.51061302073058e-05,
|
|
"loss": 5.1264,
|
|
"mean_token_accuracy": 0.2249024674296379,
|
|
"num_tokens": 16047003.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 6.039553260803222,
|
|
"epoch": 8.151168572849329,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 9.487183820496664e-05,
|
|
"loss": 4.999,
|
|
"mean_token_accuracy": 0.2408556416630745,
|
|
"num_tokens": 16056613.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.908653640747071,
|
|
"epoch": 8.156141223272003,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.463808885799582e-05,
|
|
"loss": 4.8925,
|
|
"mean_token_accuracy": 0.24619029313325883,
|
|
"num_tokens": 16066121.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.906688165664673,
|
|
"epoch": 8.16111387369468,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.440488287058819e-05,
|
|
"loss": 4.8445,
|
|
"mean_token_accuracy": 0.24742488861083983,
|
|
"num_tokens": 16076644.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 6.086437654495239,
|
|
"epoch": 8.166086524117354,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.417222094530181e-05,
|
|
"loss": 5.0938,
|
|
"mean_token_accuracy": 0.21673329770565034,
|
|
"num_tokens": 16085384.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 6.099312448501587,
|
|
"epoch": 8.17105917454003,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.39401037830556e-05,
|
|
"loss": 5.0952,
|
|
"mean_token_accuracy": 0.2256269708275795,
|
|
"num_tokens": 16094669.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 6.153309869766235,
|
|
"epoch": 8.176031824962704,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.370853208312732e-05,
|
|
"loss": 5.1182,
|
|
"mean_token_accuracy": 0.22587994635105133,
|
|
"num_tokens": 16104756.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 6.076687240600586,
|
|
"epoch": 8.18100447538538,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.347750654315165e-05,
|
|
"loss": 5.116,
|
|
"mean_token_accuracy": 0.22984861731529235,
|
|
"num_tokens": 16114103.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 6.061831903457642,
|
|
"epoch": 8.185977125808055,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.324702785911762e-05,
|
|
"loss": 5.0193,
|
|
"mean_token_accuracy": 0.22795597910881044,
|
|
"num_tokens": 16123362.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.887835359573364,
|
|
"epoch": 8.190949776230731,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.301709672536701e-05,
|
|
"loss": 4.883,
|
|
"mean_token_accuracy": 0.24873293489217757,
|
|
"num_tokens": 16132296.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 6.05796537399292,
|
|
"epoch": 8.195922426653405,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 9.278771383459189e-05,
|
|
"loss": 5.0841,
|
|
"mean_token_accuracy": 0.2255633741617203,
|
|
"num_tokens": 16141871.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 6.04565110206604,
|
|
"epoch": 8.200895077076082,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.255887987783287e-05,
|
|
"loss": 5.1294,
|
|
"mean_token_accuracy": 0.2214575618505478,
|
|
"num_tokens": 16151514.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 6.0626696109771725,
|
|
"epoch": 8.205867727498758,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.233059554447674e-05,
|
|
"loss": 5.0681,
|
|
"mean_token_accuracy": 0.22821395993232726,
|
|
"num_tokens": 16160993.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 6.056392097473145,
|
|
"epoch": 8.210840377921432,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.210286152225433e-05,
|
|
"loss": 5.0539,
|
|
"mean_token_accuracy": 0.23004778772592543,
|
|
"num_tokens": 16170444.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 6.116758060455322,
|
|
"epoch": 8.215813028344108,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 9.187567849723889e-05,
|
|
"loss": 5.1714,
|
|
"mean_token_accuracy": 0.20830632299184798,
|
|
"num_tokens": 16180137.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.865674686431885,
|
|
"epoch": 8.220785678766783,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.164904715384363e-05,
|
|
"loss": 4.7972,
|
|
"mean_token_accuracy": 0.2587510973215103,
|
|
"num_tokens": 16189536.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.859357452392578,
|
|
"epoch": 8.225758329189459,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.142296817481963e-05,
|
|
"loss": 4.8035,
|
|
"mean_token_accuracy": 0.2625247597694397,
|
|
"num_tokens": 16199762.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 6.050108814239502,
|
|
"epoch": 8.230730979612133,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.11974422412541e-05,
|
|
"loss": 5.0801,
|
|
"mean_token_accuracy": 0.22961027175188065,
|
|
"num_tokens": 16209284.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 6.071900463104248,
|
|
"epoch": 8.23570363003481,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.097247003256801e-05,
|
|
"loss": 5.0116,
|
|
"mean_token_accuracy": 0.23252638280391694,
|
|
"num_tokens": 16219150.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 6.045991134643555,
|
|
"epoch": 8.240676280457484,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.074805222651418e-05,
|
|
"loss": 4.9828,
|
|
"mean_token_accuracy": 0.24152518361806868,
|
|
"num_tokens": 16228624.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 6.147127962112426,
|
|
"epoch": 8.24564893088016,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.052418949917543e-05,
|
|
"loss": 5.1443,
|
|
"mean_token_accuracy": 0.22039050459861756,
|
|
"num_tokens": 16238274.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 6.055439281463623,
|
|
"epoch": 8.250621581302834,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.030088252496199e-05,
|
|
"loss": 5.1214,
|
|
"mean_token_accuracy": 0.22349278330802919,
|
|
"num_tokens": 16247303.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 6.0280543804168705,
|
|
"epoch": 8.25559423172551,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.007813197661016e-05,
|
|
"loss": 4.9259,
|
|
"mean_token_accuracy": 0.24405514001846312,
|
|
"num_tokens": 16257324.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 6.026740312576294,
|
|
"epoch": 8.260566882148185,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.985593852517979e-05,
|
|
"loss": 5.0342,
|
|
"mean_token_accuracy": 0.23256106674671173,
|
|
"num_tokens": 16267784.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 6.031046152114868,
|
|
"epoch": 8.26553953257086,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.963430284005235e-05,
|
|
"loss": 5.0034,
|
|
"mean_token_accuracy": 0.2314118802547455,
|
|
"num_tokens": 16277706.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.936210775375367,
|
|
"epoch": 8.270512182993535,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.941322558892918e-05,
|
|
"loss": 4.9672,
|
|
"mean_token_accuracy": 0.24114347845315934,
|
|
"num_tokens": 16288389.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 6.085309457778931,
|
|
"epoch": 8.275484833416211,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.919270743782903e-05,
|
|
"loss": 5.123,
|
|
"mean_token_accuracy": 0.22239165008068085,
|
|
"num_tokens": 16298343.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 6.0981837749481205,
|
|
"epoch": 8.280457483838886,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.897274905108653e-05,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.22493064999580384,
|
|
"num_tokens": 16307547.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 6.0365794658660885,
|
|
"epoch": 8.285430134261562,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.87533510913499e-05,
|
|
"loss": 5.0595,
|
|
"mean_token_accuracy": 0.22631748765707016,
|
|
"num_tokens": 16317378.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 6.108015680313111,
|
|
"epoch": 8.290402784684236,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.853451421957887e-05,
|
|
"loss": 5.0389,
|
|
"mean_token_accuracy": 0.23070097118616104,
|
|
"num_tokens": 16327085.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 6.071615743637085,
|
|
"epoch": 8.295375435106912,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.831623909504304e-05,
|
|
"loss": 5.1189,
|
|
"mean_token_accuracy": 0.22351790219545364,
|
|
"num_tokens": 16337401.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 6.0374504089355465,
|
|
"epoch": 8.300348085529587,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.809852637531951e-05,
|
|
"loss": 5.0974,
|
|
"mean_token_accuracy": 0.2266927793622017,
|
|
"num_tokens": 16347141.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 6.0519976139068605,
|
|
"epoch": 8.305320735952263,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.788137671629121e-05,
|
|
"loss": 5.0887,
|
|
"mean_token_accuracy": 0.22517900168895721,
|
|
"num_tokens": 16356418.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.9234232902526855,
|
|
"epoch": 8.310293386374937,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.766479077214466e-05,
|
|
"loss": 4.7768,
|
|
"mean_token_accuracy": 0.2580944046378136,
|
|
"num_tokens": 16365218.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 6.094912052154541,
|
|
"epoch": 8.315266036797613,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 8.744876919536832e-05,
|
|
"loss": 5.1414,
|
|
"mean_token_accuracy": 0.22950075268745423,
|
|
"num_tokens": 16374764.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 6.075074338912964,
|
|
"epoch": 8.320238687220288,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.723331263675021e-05,
|
|
"loss": 5.0756,
|
|
"mean_token_accuracy": 0.2308366686105728,
|
|
"num_tokens": 16383797.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 6.076484155654907,
|
|
"epoch": 8.325211337642964,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.701842174537627e-05,
|
|
"loss": 5.0748,
|
|
"mean_token_accuracy": 0.22540437877178193,
|
|
"num_tokens": 16394349.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 6.120758008956909,
|
|
"epoch": 8.330183988065638,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.680409716862833e-05,
|
|
"loss": 5.0776,
|
|
"mean_token_accuracy": 0.22220608741044998,
|
|
"num_tokens": 16404890.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 6.046685647964478,
|
|
"epoch": 8.335156638488314,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.659033955218221e-05,
|
|
"loss": 4.9999,
|
|
"mean_token_accuracy": 0.23605923354625702,
|
|
"num_tokens": 16414943.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 6.062542247772217,
|
|
"epoch": 8.34012928891099,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.637714954000548e-05,
|
|
"loss": 4.9835,
|
|
"mean_token_accuracy": 0.23613086938858033,
|
|
"num_tokens": 16425578.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.9554527759552,
|
|
"epoch": 8.345101939333665,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.616452777435597e-05,
|
|
"loss": 4.9126,
|
|
"mean_token_accuracy": 0.2568305715918541,
|
|
"num_tokens": 16436376.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 6.070420837402343,
|
|
"epoch": 8.35007458975634,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.595247489577939e-05,
|
|
"loss": 5.057,
|
|
"mean_token_accuracy": 0.22900198996067048,
|
|
"num_tokens": 16445708.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 6.118132400512695,
|
|
"epoch": 8.355047240179015,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.574099154310782e-05,
|
|
"loss": 5.1855,
|
|
"mean_token_accuracy": 0.2197991907596588,
|
|
"num_tokens": 16456306.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 6.046338939666748,
|
|
"epoch": 8.360019890601691,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 8.553007835345755e-05,
|
|
"loss": 5.0529,
|
|
"mean_token_accuracy": 0.22626874595880508,
|
|
"num_tokens": 16465579.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 6.148076820373535,
|
|
"epoch": 8.364992541024366,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.531973596222701e-05,
|
|
"loss": 5.158,
|
|
"mean_token_accuracy": 0.2154083341360092,
|
|
"num_tokens": 16476050.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 6.121895265579224,
|
|
"epoch": 8.369965191447042,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.510996500309531e-05,
|
|
"loss": 5.1528,
|
|
"mean_token_accuracy": 0.21992243081331253,
|
|
"num_tokens": 16485745.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 6.077825355529785,
|
|
"epoch": 8.374937841869716,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.490076610801983e-05,
|
|
"loss": 5.084,
|
|
"mean_token_accuracy": 0.22568577975034715,
|
|
"num_tokens": 16495121.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 6.06300687789917,
|
|
"epoch": 8.379910492292392,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.46921399072346e-05,
|
|
"loss": 5.0891,
|
|
"mean_token_accuracy": 0.2304096668958664,
|
|
"num_tokens": 16505782.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 6.10689377784729,
|
|
"epoch": 8.384883142715067,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.448408702924846e-05,
|
|
"loss": 5.1526,
|
|
"mean_token_accuracy": 0.22304088324308396,
|
|
"num_tokens": 16516068.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.968610143661499,
|
|
"epoch": 8.389855793137743,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.427660810084292e-05,
|
|
"loss": 4.922,
|
|
"mean_token_accuracy": 0.2482813522219658,
|
|
"num_tokens": 16525809.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 6.095130109786988,
|
|
"epoch": 8.394828443560417,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.406970374707048e-05,
|
|
"loss": 5.093,
|
|
"mean_token_accuracy": 0.22460667937994003,
|
|
"num_tokens": 16535823.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 6.067734146118164,
|
|
"epoch": 8.399801093983093,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.386337459125268e-05,
|
|
"loss": 5.0819,
|
|
"mean_token_accuracy": 0.22689371109008788,
|
|
"num_tokens": 16545889.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.965697431564331,
|
|
"epoch": 8.404773744405768,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.365762125497814e-05,
|
|
"loss": 4.9659,
|
|
"mean_token_accuracy": 0.2415914788842201,
|
|
"num_tokens": 16555945.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 6.0539772510528564,
|
|
"epoch": 8.409746394828444,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 8.345244435810084e-05,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.227622589468956,
|
|
"num_tokens": 16566417.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.962084197998047,
|
|
"epoch": 8.414719045251118,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 8.324784451873813e-05,
|
|
"loss": 4.8382,
|
|
"mean_token_accuracy": 0.2520796462893486,
|
|
"num_tokens": 16576524.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 6.056379175186157,
|
|
"epoch": 8.419691695673794,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.304382235326893e-05,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.22266504019498826,
|
|
"num_tokens": 16586512.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 6.024080467224121,
|
|
"epoch": 8.424664346096469,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 8.284037847633189e-05,
|
|
"loss": 5.0979,
|
|
"mean_token_accuracy": 0.22875019609928132,
|
|
"num_tokens": 16596866.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 6.04751181602478,
|
|
"epoch": 8.429636996519145,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.263751350082345e-05,
|
|
"loss": 5.0766,
|
|
"mean_token_accuracy": 0.23186299055814744,
|
|
"num_tokens": 16606506.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 6.0908979892730715,
|
|
"epoch": 8.43460964694182,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.243522803789597e-05,
|
|
"loss": 5.0795,
|
|
"mean_token_accuracy": 0.2278062492609024,
|
|
"num_tokens": 16615836.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 6.112334537506103,
|
|
"epoch": 8.439582297364495,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.22335226969562e-05,
|
|
"loss": 5.0954,
|
|
"mean_token_accuracy": 0.22952943444252014,
|
|
"num_tokens": 16624601.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.9841203689575195,
|
|
"epoch": 8.44455494778717,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.203239808566296e-05,
|
|
"loss": 4.9117,
|
|
"mean_token_accuracy": 0.2431499630212784,
|
|
"num_tokens": 16635389.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 6.12209849357605,
|
|
"epoch": 8.449527598209846,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.183185480992575e-05,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.21691016852855682,
|
|
"num_tokens": 16645944.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 6.083438062667847,
|
|
"epoch": 8.45450024863252,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.163189347390259e-05,
|
|
"loss": 5.1701,
|
|
"mean_token_accuracy": 0.21100667864084244,
|
|
"num_tokens": 16656416.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 6.065715265274048,
|
|
"epoch": 8.459472899055196,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 8.14325146799984e-05,
|
|
"loss": 5.1064,
|
|
"mean_token_accuracy": 0.22572797834873198,
|
|
"num_tokens": 16667932.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 6.071225357055664,
|
|
"epoch": 8.464445549477873,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.123371902886325e-05,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.2146118849515915,
|
|
"num_tokens": 16678769.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 6.0038923740386965,
|
|
"epoch": 8.469418199900547,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.103550711939018e-05,
|
|
"loss": 5.0349,
|
|
"mean_token_accuracy": 0.23087847381830215,
|
|
"num_tokens": 16688548.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 6.045148181915283,
|
|
"epoch": 8.474390850323223,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.08378795487139e-05,
|
|
"loss": 5.0774,
|
|
"mean_token_accuracy": 0.22579897046089173,
|
|
"num_tokens": 16698670.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 6.113581752777099,
|
|
"epoch": 8.479363500745897,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.064083691220852e-05,
|
|
"loss": 5.0583,
|
|
"mean_token_accuracy": 0.23144163191318512,
|
|
"num_tokens": 16708450.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 6.166110467910767,
|
|
"epoch": 8.484336151168574,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.044437980348619e-05,
|
|
"loss": 5.1424,
|
|
"mean_token_accuracy": 0.2230359598994255,
|
|
"num_tokens": 16718124.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.963276815414429,
|
|
"epoch": 8.489308801591248,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 8.02485088143949e-05,
|
|
"loss": 4.9574,
|
|
"mean_token_accuracy": 0.24565739184617996,
|
|
"num_tokens": 16728919.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 6.072273778915405,
|
|
"epoch": 8.494281452013924,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.005322453501692e-05,
|
|
"loss": 5.0612,
|
|
"mean_token_accuracy": 0.22857411354780197,
|
|
"num_tokens": 16738365.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 6.067027378082275,
|
|
"epoch": 8.499254102436598,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.985852755366717e-05,
|
|
"loss": 4.9479,
|
|
"mean_token_accuracy": 0.2484827980399132,
|
|
"num_tokens": 16747158.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 6.0174517154693605,
|
|
"epoch": 8.504226752859275,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.966441845689102e-05,
|
|
"loss": 5.0131,
|
|
"mean_token_accuracy": 0.22946599125862122,
|
|
"num_tokens": 16755636.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 6.092843961715698,
|
|
"epoch": 8.509199403281949,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.947089782946295e-05,
|
|
"loss": 5.2255,
|
|
"mean_token_accuracy": 0.217678302526474,
|
|
"num_tokens": 16764932.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 6.1361494064331055,
|
|
"epoch": 8.514172053704625,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.927796625438466e-05,
|
|
"loss": 5.0677,
|
|
"mean_token_accuracy": 0.2289421319961548,
|
|
"num_tokens": 16776889.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.97844877243042,
|
|
"epoch": 8.5191447041273,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.908562431288303e-05,
|
|
"loss": 5.0039,
|
|
"mean_token_accuracy": 0.238939169049263,
|
|
"num_tokens": 16787855.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 6.133350896835327,
|
|
"epoch": 8.524117354549976,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.889387258440884e-05,
|
|
"loss": 5.1004,
|
|
"mean_token_accuracy": 0.22410138994455336,
|
|
"num_tokens": 16796907.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 6.154951572418213,
|
|
"epoch": 8.52909000497265,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.870271164663463e-05,
|
|
"loss": 5.114,
|
|
"mean_token_accuracy": 0.2254410520195961,
|
|
"num_tokens": 16806778.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 6.092426490783692,
|
|
"epoch": 8.534062655395326,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.851214207545322e-05,
|
|
"loss": 5.1534,
|
|
"mean_token_accuracy": 0.22013937681913376,
|
|
"num_tokens": 16816249.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 6.107940149307251,
|
|
"epoch": 8.539035305818,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.832216444497589e-05,
|
|
"loss": 5.1256,
|
|
"mean_token_accuracy": 0.22767855674028398,
|
|
"num_tokens": 16825726.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 6.058207893371582,
|
|
"epoch": 8.544007956240677,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.813277932753053e-05,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.22190484553575515,
|
|
"num_tokens": 16834833.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 6.077880382537842,
|
|
"epoch": 8.548980606663351,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.794398729366006e-05,
|
|
"loss": 5.0761,
|
|
"mean_token_accuracy": 0.22523839622735978,
|
|
"num_tokens": 16843414.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 6.166740608215332,
|
|
"epoch": 8.553953257086027,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.775578891212076e-05,
|
|
"loss": 5.1613,
|
|
"mean_token_accuracy": 0.21516168862581253,
|
|
"num_tokens": 16853602.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 6.081193256378174,
|
|
"epoch": 8.558925907508701,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.75681847498803e-05,
|
|
"loss": 5.0577,
|
|
"mean_token_accuracy": 0.22817078232765198,
|
|
"num_tokens": 16862845.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 6.077098703384399,
|
|
"epoch": 8.563898557931378,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.73811753721164e-05,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.23957579135894774,
|
|
"num_tokens": 16872475.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 6.049982213973999,
|
|
"epoch": 8.568871208354052,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.719476134221474e-05,
|
|
"loss": 5.0296,
|
|
"mean_token_accuracy": 0.2407066375017166,
|
|
"num_tokens": 16882161.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 6.03905987739563,
|
|
"epoch": 8.573843858776728,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.70089432217676e-05,
|
|
"loss": 5.0104,
|
|
"mean_token_accuracy": 0.24186116456985474,
|
|
"num_tokens": 16891607.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 6.089488172531128,
|
|
"epoch": 8.578816509199402,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.682372157057196e-05,
|
|
"loss": 5.0815,
|
|
"mean_token_accuracy": 0.22365832775831224,
|
|
"num_tokens": 16900750.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 6.0744904518127445,
|
|
"epoch": 8.583789159622079,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.663909694662783e-05,
|
|
"loss": 5.0973,
|
|
"mean_token_accuracy": 0.22188807129859925,
|
|
"num_tokens": 16910441.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 6.056908845901489,
|
|
"epoch": 8.588761810044755,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.64550699061367e-05,
|
|
"loss": 5.0789,
|
|
"mean_token_accuracy": 0.22560325264930725,
|
|
"num_tokens": 16919331.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 6.053385305404663,
|
|
"epoch": 8.593734460467429,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.627164100349967e-05,
|
|
"loss": 5.1164,
|
|
"mean_token_accuracy": 0.22388401329517366,
|
|
"num_tokens": 16930401.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 6.158016586303711,
|
|
"epoch": 8.598707110890105,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.608881079131604e-05,
|
|
"loss": 5.0911,
|
|
"mean_token_accuracy": 0.22067286521196366,
|
|
"num_tokens": 16939552.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 6.032980728149414,
|
|
"epoch": 8.60367976131278,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.59065798203813e-05,
|
|
"loss": 5.0665,
|
|
"mean_token_accuracy": 0.22334950119256974,
|
|
"num_tokens": 16950022.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 6.019799613952637,
|
|
"epoch": 8.608652411735456,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.572494863968573e-05,
|
|
"loss": 4.969,
|
|
"mean_token_accuracy": 0.24503548443317413,
|
|
"num_tokens": 16959583.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.96540641784668,
|
|
"epoch": 8.61362506215813,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.55439177964128e-05,
|
|
"loss": 4.9323,
|
|
"mean_token_accuracy": 0.23636542856693268,
|
|
"num_tokens": 16968541.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.921866750717163,
|
|
"epoch": 8.618597712580806,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.536348783593727e-05,
|
|
"loss": 4.8955,
|
|
"mean_token_accuracy": 0.2503947988152504,
|
|
"num_tokens": 16978131.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 6.12747163772583,
|
|
"epoch": 8.62357036300348,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.518365930182368e-05,
|
|
"loss": 5.1339,
|
|
"mean_token_accuracy": 0.2198960691690445,
|
|
"num_tokens": 16989160.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 6.121262836456299,
|
|
"epoch": 8.628543013426157,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.500443273582478e-05,
|
|
"loss": 5.0662,
|
|
"mean_token_accuracy": 0.23210011720657348,
|
|
"num_tokens": 16998771.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 6.086851596832275,
|
|
"epoch": 8.633515663848831,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.482580867787974e-05,
|
|
"loss": 5.0624,
|
|
"mean_token_accuracy": 0.22406096905469894,
|
|
"num_tokens": 17007278.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.950457286834717,
|
|
"epoch": 8.638488314271507,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.464778766611266e-05,
|
|
"loss": 4.9147,
|
|
"mean_token_accuracy": 0.23844432234764099,
|
|
"num_tokens": 17018472.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 6.111311340332032,
|
|
"epoch": 8.643460964694182,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.447037023683095e-05,
|
|
"loss": 5.1266,
|
|
"mean_token_accuracy": 0.22320729047060012,
|
|
"num_tokens": 17028542.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 6.093828344345093,
|
|
"epoch": 8.648433615116858,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.429355692452355e-05,
|
|
"loss": 5.0559,
|
|
"mean_token_accuracy": 0.2295610561966896,
|
|
"num_tokens": 17037226.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 6.064908695220947,
|
|
"epoch": 8.653406265539532,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.411734826185953e-05,
|
|
"loss": 5.1354,
|
|
"mean_token_accuracy": 0.2202439695596695,
|
|
"num_tokens": 17046949.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 6.09107027053833,
|
|
"epoch": 8.658378915962208,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.394174477968632e-05,
|
|
"loss": 5.1313,
|
|
"mean_token_accuracy": 0.2288019835948944,
|
|
"num_tokens": 17056753.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 6.042801237106323,
|
|
"epoch": 8.663351566384883,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 7.376674700702817e-05,
|
|
"loss": 5.0649,
|
|
"mean_token_accuracy": 0.23041289448738098,
|
|
"num_tokens": 17065859.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 6.12708477973938,
|
|
"epoch": 8.668324216807559,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.359235547108469e-05,
|
|
"loss": 5.1397,
|
|
"mean_token_accuracy": 0.22472907304763795,
|
|
"num_tokens": 17075699.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 6.089188289642334,
|
|
"epoch": 8.673296867230233,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.341857069722897e-05,
|
|
"loss": 5.1086,
|
|
"mean_token_accuracy": 0.22664082199335098,
|
|
"num_tokens": 17085813.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 6.025517177581787,
|
|
"epoch": 8.67826951765291,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.324539320900632e-05,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.23410575985908508,
|
|
"num_tokens": 17095641.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 6.123284006118775,
|
|
"epoch": 8.683242168075584,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.307282352813237e-05,
|
|
"loss": 5.1626,
|
|
"mean_token_accuracy": 0.2178654119372368,
|
|
"num_tokens": 17105776.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 6.127776336669922,
|
|
"epoch": 8.68821481849826,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.290086217449176e-05,
|
|
"loss": 5.1635,
|
|
"mean_token_accuracy": 0.22049986720085143,
|
|
"num_tokens": 17115157.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 6.093261957168579,
|
|
"epoch": 8.693187468920934,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.272950966613659e-05,
|
|
"loss": 5.0774,
|
|
"mean_token_accuracy": 0.2242334231734276,
|
|
"num_tokens": 17124560.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 6.088965177536011,
|
|
"epoch": 8.69816011934361,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.255876651928453e-05,
|
|
"loss": 5.0034,
|
|
"mean_token_accuracy": 0.2293031707406044,
|
|
"num_tokens": 17134235.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.936453151702881,
|
|
"epoch": 8.703132769766285,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.238863324831762e-05,
|
|
"loss": 4.9303,
|
|
"mean_token_accuracy": 0.24751580506563187,
|
|
"num_tokens": 17143888.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 6.02718505859375,
|
|
"epoch": 8.70810542018896,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.221911036578053e-05,
|
|
"loss": 5.0218,
|
|
"mean_token_accuracy": 0.233767069876194,
|
|
"num_tokens": 17152766.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 6.0859230041503904,
|
|
"epoch": 8.713078070611637,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.205019838237916e-05,
|
|
"loss": 5.0911,
|
|
"mean_token_accuracy": 0.21861542910337448,
|
|
"num_tokens": 17162489.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 6.144061756134033,
|
|
"epoch": 8.718050721034311,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.188189780697885e-05,
|
|
"loss": 5.1321,
|
|
"mean_token_accuracy": 0.23180071711540223,
|
|
"num_tokens": 17172976.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 6.018468856811523,
|
|
"epoch": 8.723023371456987,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.171420914660308e-05,
|
|
"loss": 5.0316,
|
|
"mean_token_accuracy": 0.2321193754673004,
|
|
"num_tokens": 17182687.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 6.0772205829620365,
|
|
"epoch": 8.727996021879662,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.154713290643195e-05,
|
|
"loss": 5.0296,
|
|
"mean_token_accuracy": 0.22682259231805801,
|
|
"num_tokens": 17192354.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.945929718017578,
|
|
"epoch": 8.732968672302338,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.13806695898006e-05,
|
|
"loss": 4.8995,
|
|
"mean_token_accuracy": 0.24974654316902162,
|
|
"num_tokens": 17202805.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 6.059028244018554,
|
|
"epoch": 8.737941322725012,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.121481969819739e-05,
|
|
"loss": 5.0346,
|
|
"mean_token_accuracy": 0.23177895098924636,
|
|
"num_tokens": 17212805.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 6.095387935638428,
|
|
"epoch": 8.742913973147688,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.104958373126306e-05,
|
|
"loss": 5.062,
|
|
"mean_token_accuracy": 0.22986815571784974,
|
|
"num_tokens": 17222797.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 6.075600242614746,
|
|
"epoch": 8.747886623570363,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.08849621867885e-05,
|
|
"loss": 5.1172,
|
|
"mean_token_accuracy": 0.22972231656312941,
|
|
"num_tokens": 17233120.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 6.086778068542481,
|
|
"epoch": 8.752859273993039,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.072095556071381e-05,
|
|
"loss": 5.083,
|
|
"mean_token_accuracy": 0.22640911340713502,
|
|
"num_tokens": 17241234.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 6.1235157489776615,
|
|
"epoch": 8.757831924415713,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.055756434712653e-05,
|
|
"loss": 5.1446,
|
|
"mean_token_accuracy": 0.22453143894672395,
|
|
"num_tokens": 17250502.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 6.142209768295288,
|
|
"epoch": 8.76280457483839,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.039478903826012e-05,
|
|
"loss": 5.234,
|
|
"mean_token_accuracy": 0.21119848340749742,
|
|
"num_tokens": 17260400.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 6.0296765804290775,
|
|
"epoch": 8.767777225261064,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.023263012449269e-05,
|
|
"loss": 4.9896,
|
|
"mean_token_accuracy": 0.2334146037697792,
|
|
"num_tokens": 17269855.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 6.1193784236907955,
|
|
"epoch": 8.77274987568374,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.007108809434527e-05,
|
|
"loss": 5.0992,
|
|
"mean_token_accuracy": 0.222120463848114,
|
|
"num_tokens": 17280005.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.994942903518677,
|
|
"epoch": 8.777722526106414,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.99101634344805e-05,
|
|
"loss": 5.0076,
|
|
"mean_token_accuracy": 0.2367738217115402,
|
|
"num_tokens": 17289929.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 6.091522216796875,
|
|
"epoch": 8.78269517652909,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.974985662970119e-05,
|
|
"loss": 5.0581,
|
|
"mean_token_accuracy": 0.2355152651667595,
|
|
"num_tokens": 17299572.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.960417032241821,
|
|
"epoch": 8.787667826951765,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.959016816294872e-05,
|
|
"loss": 4.9606,
|
|
"mean_token_accuracy": 0.24318485707044601,
|
|
"num_tokens": 17308905.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 6.07923436164856,
|
|
"epoch": 8.79264047737444,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 6.943109851530168e-05,
|
|
"loss": 5.1102,
|
|
"mean_token_accuracy": 0.2286006197333336,
|
|
"num_tokens": 17320145.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 6.084366989135742,
|
|
"epoch": 8.797613127797115,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.927264816597448e-05,
|
|
"loss": 5.0556,
|
|
"mean_token_accuracy": 0.23269785344600677,
|
|
"num_tokens": 17330189.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 6.129771327972412,
|
|
"epoch": 8.802585778219791,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.911481759231566e-05,
|
|
"loss": 5.1093,
|
|
"mean_token_accuracy": 0.22310526072978973,
|
|
"num_tokens": 17339154.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 6.0696367740631105,
|
|
"epoch": 8.807558428642466,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 6.895760726980681e-05,
|
|
"loss": 5.1209,
|
|
"mean_token_accuracy": 0.22580073773860931,
|
|
"num_tokens": 17348611.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 6.109683513641357,
|
|
"epoch": 8.812531079065142,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.880101767206079e-05,
|
|
"loss": 5.1452,
|
|
"mean_token_accuracy": 0.22322923094034194,
|
|
"num_tokens": 17358572.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.910505867004394,
|
|
"epoch": 8.817503729487816,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.864504927082051e-05,
|
|
"loss": 4.8979,
|
|
"mean_token_accuracy": 0.2545802965760231,
|
|
"num_tokens": 17368464.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 6.114032030105591,
|
|
"epoch": 8.822476379910492,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.848970253595759e-05,
|
|
"loss": 5.0822,
|
|
"mean_token_accuracy": 0.2227763295173645,
|
|
"num_tokens": 17377326.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 6.074749374389649,
|
|
"epoch": 8.827449030333167,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.833497793547059e-05,
|
|
"loss": 5.1115,
|
|
"mean_token_accuracy": 0.22889809161424637,
|
|
"num_tokens": 17387551.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 6.128981113433838,
|
|
"epoch": 8.832421680755843,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.818087593548391e-05,
|
|
"loss": 5.1958,
|
|
"mean_token_accuracy": 0.21451227813959123,
|
|
"num_tokens": 17398167.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 6.064513492584228,
|
|
"epoch": 8.837394331178519,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 6.802739700024643e-05,
|
|
"loss": 5.0639,
|
|
"mean_token_accuracy": 0.2253905400633812,
|
|
"num_tokens": 17408640.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.978671932220459,
|
|
"epoch": 8.842366981601193,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.787454159212974e-05,
|
|
"loss": 4.9058,
|
|
"mean_token_accuracy": 0.2456461265683174,
|
|
"num_tokens": 17418296.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.941046237945557,
|
|
"epoch": 8.84733963202387,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.772231017162725e-05,
|
|
"loss": 4.9166,
|
|
"mean_token_accuracy": 0.25066306442022324,
|
|
"num_tokens": 17428714.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 6.076188850402832,
|
|
"epoch": 8.852312282446544,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.75707031973523e-05,
|
|
"loss": 5.2208,
|
|
"mean_token_accuracy": 0.21417063027620314,
|
|
"num_tokens": 17439261.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 6.108159399032592,
|
|
"epoch": 8.85728493286922,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.741972112603719e-05,
|
|
"loss": 5.1411,
|
|
"mean_token_accuracy": 0.21693692356348038,
|
|
"num_tokens": 17449352.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 6.044029331207275,
|
|
"epoch": 8.862257583291894,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.726936441253155e-05,
|
|
"loss": 5.0824,
|
|
"mean_token_accuracy": 0.2241876542568207,
|
|
"num_tokens": 17460019.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 6.06287784576416,
|
|
"epoch": 8.86723023371457,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.711963350980106e-05,
|
|
"loss": 5.0741,
|
|
"mean_token_accuracy": 0.2279302582144737,
|
|
"num_tokens": 17470014.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 6.021813583374024,
|
|
"epoch": 8.872202884137245,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.697052886892614e-05,
|
|
"loss": 5.0561,
|
|
"mean_token_accuracy": 0.2383757010102272,
|
|
"num_tokens": 17481126.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 6.137328147888184,
|
|
"epoch": 8.877175534559921,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.682205093910044e-05,
|
|
"loss": 5.1427,
|
|
"mean_token_accuracy": 0.22543676495552062,
|
|
"num_tokens": 17492114.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 6.086642456054688,
|
|
"epoch": 8.882148184982595,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.667420016762963e-05,
|
|
"loss": 5.1585,
|
|
"mean_token_accuracy": 0.21796324700117112,
|
|
"num_tokens": 17501902.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 6.043990135192871,
|
|
"epoch": 8.887120835405272,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 6.652697699993e-05,
|
|
"loss": 5.075,
|
|
"mean_token_accuracy": 0.22718257308006287,
|
|
"num_tokens": 17512321.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 6.127042627334594,
|
|
"epoch": 8.892093485827946,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.638038187952705e-05,
|
|
"loss": 5.1595,
|
|
"mean_token_accuracy": 0.21753038018941878,
|
|
"num_tokens": 17521681.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 6.0526939868927006,
|
|
"epoch": 8.897066136250622,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.623441524805433e-05,
|
|
"loss": 4.9722,
|
|
"mean_token_accuracy": 0.24593328833580017,
|
|
"num_tokens": 17530378.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 6.157143449783325,
|
|
"epoch": 8.902038786673296,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 6.608907754525189e-05,
|
|
"loss": 5.1516,
|
|
"mean_token_accuracy": 0.2263398438692093,
|
|
"num_tokens": 17540310.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 6.085373878479004,
|
|
"epoch": 8.907011437095973,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.594436920896511e-05,
|
|
"loss": 5.1046,
|
|
"mean_token_accuracy": 0.23018069118261336,
|
|
"num_tokens": 17550413.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 6.0705818176269535,
|
|
"epoch": 8.911984087518647,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.580029067514346e-05,
|
|
"loss": 5.1133,
|
|
"mean_token_accuracy": 0.22671385407447814,
|
|
"num_tokens": 17559627.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.985567855834961,
|
|
"epoch": 8.916956737941323,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.565684237783881e-05,
|
|
"loss": 5.0128,
|
|
"mean_token_accuracy": 0.23453458100557328,
|
|
"num_tokens": 17569562.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 6.079870128631592,
|
|
"epoch": 8.921929388363997,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.551402474920458e-05,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.22050649225711821,
|
|
"num_tokens": 17579763.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 6.02824649810791,
|
|
"epoch": 8.926902038786674,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.537183821949412e-05,
|
|
"loss": 5.0337,
|
|
"mean_token_accuracy": 0.23316333889961244,
|
|
"num_tokens": 17589525.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.97695517539978,
|
|
"epoch": 8.931874689209348,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 6.52302832170596e-05,
|
|
"loss": 4.9645,
|
|
"mean_token_accuracy": 0.24301348477602006,
|
|
"num_tokens": 17599647.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 6.091416454315185,
|
|
"epoch": 8.936847339632024,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 6.508936016835063e-05,
|
|
"loss": 5.0572,
|
|
"mean_token_accuracy": 0.2366153195500374,
|
|
"num_tokens": 17608609.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 6.118017673492432,
|
|
"epoch": 8.941819990054698,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.494906949791296e-05,
|
|
"loss": 5.1767,
|
|
"mean_token_accuracy": 0.21379632502794266,
|
|
"num_tokens": 17618529.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 6.095211839675903,
|
|
"epoch": 8.946792640477375,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.480941162838724e-05,
|
|
"loss": 5.1658,
|
|
"mean_token_accuracy": 0.2207474485039711,
|
|
"num_tokens": 17627547.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 8.946792640477375,
|
|
"eval_entropy": 5.9398862150956955,
|
|
"eval_loss": 6.449407577514648,
|
|
"eval_mean_token_accuracy": 0.15090311813052348,
|
|
"eval_num_tokens": 17627547.0,
|
|
"eval_runtime": 1.9265,
|
|
"eval_samples_per_second": 1798.571,
|
|
"eval_steps_per_second": 225.276,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 6.058997344970703,
|
|
"epoch": 8.951765290900049,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.46703869805078e-05,
|
|
"loss": 5.1192,
|
|
"mean_token_accuracy": 0.22088052779436113,
|
|
"num_tokens": 17638485.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 6.082899045944214,
|
|
"epoch": 8.956737941322725,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.453199597310119e-05,
|
|
"loss": 5.0487,
|
|
"mean_token_accuracy": 0.2317546173930168,
|
|
"num_tokens": 17648088.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 6.060591459274292,
|
|
"epoch": 8.961710591745401,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.439423902308521e-05,
|
|
"loss": 4.9729,
|
|
"mean_token_accuracy": 0.23445261418819427,
|
|
"num_tokens": 17656681.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 6.086190557479858,
|
|
"epoch": 8.966683242168076,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.425711654546736e-05,
|
|
"loss": 5.1282,
|
|
"mean_token_accuracy": 0.22915284037590028,
|
|
"num_tokens": 17665837.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 6.091594791412353,
|
|
"epoch": 8.971655892590752,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 6.41206289533438e-05,
|
|
"loss": 5.0584,
|
|
"mean_token_accuracy": 0.22852095514535903,
|
|
"num_tokens": 17674572.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 6.065656042098999,
|
|
"epoch": 8.976628543013426,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.398477665789803e-05,
|
|
"loss": 5.1657,
|
|
"mean_token_accuracy": 0.2160267412662506,
|
|
"num_tokens": 17684912.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 6.128516769409179,
|
|
"epoch": 8.981601193436102,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.384956006839961e-05,
|
|
"loss": 5.1642,
|
|
"mean_token_accuracy": 0.2163771942257881,
|
|
"num_tokens": 17694706.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 6.047650718688965,
|
|
"epoch": 8.986573843858777,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.371497959220304e-05,
|
|
"loss": 5.0374,
|
|
"mean_token_accuracy": 0.23565800935029985,
|
|
"num_tokens": 17704798.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 6.116561794281006,
|
|
"epoch": 8.991546494281453,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.358103563474632e-05,
|
|
"loss": 5.0404,
|
|
"mean_token_accuracy": 0.2241012081503868,
|
|
"num_tokens": 17714650.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 6.024302911758423,
|
|
"epoch": 8.996519144704127,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.344772859955005e-05,
|
|
"loss": 5.1097,
|
|
"mean_token_accuracy": 0.23307857364416124,
|
|
"num_tokens": 17724636.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.854769759707981,
|
|
"epoch": 9.000994530084535,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.331505888821592e-05,
|
|
"loss": 4.821,
|
|
"mean_token_accuracy": 0.2562614629666011,
|
|
"num_tokens": 17732681.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.992326068878174,
|
|
"epoch": 9.005967180507211,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.318302690042557e-05,
|
|
"loss": 4.8999,
|
|
"mean_token_accuracy": 0.2486682265996933,
|
|
"num_tokens": 17742979.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 6.165087032318115,
|
|
"epoch": 9.010939830929885,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.305163303393955e-05,
|
|
"loss": 5.2175,
|
|
"mean_token_accuracy": 0.2158197432756424,
|
|
"num_tokens": 17752686.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.939579582214355,
|
|
"epoch": 9.015912481352562,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.292087768459597e-05,
|
|
"loss": 4.8308,
|
|
"mean_token_accuracy": 0.24863443225622178,
|
|
"num_tokens": 17762750.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 6.067292213439941,
|
|
"epoch": 9.020885131775236,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.279076124630922e-05,
|
|
"loss": 4.9799,
|
|
"mean_token_accuracy": 0.23500341027975083,
|
|
"num_tokens": 17771324.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 6.0928973197937015,
|
|
"epoch": 9.025857782197912,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 6.266128411106915e-05,
|
|
"loss": 5.0796,
|
|
"mean_token_accuracy": 0.22846223562955856,
|
|
"num_tokens": 17780654.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 6.043622398376465,
|
|
"epoch": 9.030830432620586,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.253244666893933e-05,
|
|
"loss": 5.0274,
|
|
"mean_token_accuracy": 0.23014433383941652,
|
|
"num_tokens": 17789420.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 6.045699882507324,
|
|
"epoch": 9.035803083043263,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 6.240424930805648e-05,
|
|
"loss": 5.0695,
|
|
"mean_token_accuracy": 0.22479693442583085,
|
|
"num_tokens": 17798630.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 6.087038230895996,
|
|
"epoch": 9.040775733465937,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.227669241462878e-05,
|
|
"loss": 5.0613,
|
|
"mean_token_accuracy": 0.2253473624587059,
|
|
"num_tokens": 17808707.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 6.115950584411621,
|
|
"epoch": 9.045748383888613,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.214977637293502e-05,
|
|
"loss": 5.0688,
|
|
"mean_token_accuracy": 0.22024950087070466,
|
|
"num_tokens": 17818644.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 6.063990592956543,
|
|
"epoch": 9.050721034311287,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 6.202350156532345e-05,
|
|
"loss": 5.0918,
|
|
"mean_token_accuracy": 0.22961248457431793,
|
|
"num_tokens": 17829464.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 6.105351161956787,
|
|
"epoch": 9.055693684733964,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.189786837221042e-05,
|
|
"loss": 5.0461,
|
|
"mean_token_accuracy": 0.2278224915266037,
|
|
"num_tokens": 17838055.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 6.07191309928894,
|
|
"epoch": 9.060666335156638,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.177287717207928e-05,
|
|
"loss": 5.0323,
|
|
"mean_token_accuracy": 0.23400785773992538,
|
|
"num_tokens": 17848515.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 6.074994468688965,
|
|
"epoch": 9.065638985579314,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.16485283414795e-05,
|
|
"loss": 5.0292,
|
|
"mean_token_accuracy": 0.23288363814353943,
|
|
"num_tokens": 17858181.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 6.0935498714447025,
|
|
"epoch": 9.070611636001988,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 6.152482225502518e-05,
|
|
"loss": 5.0558,
|
|
"mean_token_accuracy": 0.23113997727632524,
|
|
"num_tokens": 17868314.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 6.081724691390991,
|
|
"epoch": 9.075584286424665,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.140175928539419e-05,
|
|
"loss": 5.1271,
|
|
"mean_token_accuracy": 0.22636976689100266,
|
|
"num_tokens": 17877444.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 6.100804615020752,
|
|
"epoch": 9.080556936847339,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.127933980332685e-05,
|
|
"loss": 5.1242,
|
|
"mean_token_accuracy": 0.2207667663693428,
|
|
"num_tokens": 17887311.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 6.068193960189819,
|
|
"epoch": 9.085529587270015,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.1157564177625e-05,
|
|
"loss": 5.1239,
|
|
"mean_token_accuracy": 0.2223875716328621,
|
|
"num_tokens": 17897354.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 6.023849868774414,
|
|
"epoch": 9.09050223769269,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.103643277515075e-05,
|
|
"loss": 4.9558,
|
|
"mean_token_accuracy": 0.24610256999731064,
|
|
"num_tokens": 17906054.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 6.062192296981811,
|
|
"epoch": 9.095474888115366,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.0915945960825393e-05,
|
|
"loss": 4.9865,
|
|
"mean_token_accuracy": 0.23860390931367875,
|
|
"num_tokens": 17915851.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 6.125268602371216,
|
|
"epoch": 9.100447538538042,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.079610409762836e-05,
|
|
"loss": 5.0793,
|
|
"mean_token_accuracy": 0.22738225013017654,
|
|
"num_tokens": 17925489.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 6.127257680892944,
|
|
"epoch": 9.105420188960716,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 6.067690754659605e-05,
|
|
"loss": 5.1489,
|
|
"mean_token_accuracy": 0.22724745273590088,
|
|
"num_tokens": 17935488.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 6.115323543548584,
|
|
"epoch": 9.110392839383392,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.055835666682093e-05,
|
|
"loss": 5.1823,
|
|
"mean_token_accuracy": 0.2227105140686035,
|
|
"num_tokens": 17944626.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 6.116650581359863,
|
|
"epoch": 9.115365489806067,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.044045181545013e-05,
|
|
"loss": 5.0341,
|
|
"mean_token_accuracy": 0.2301251158118248,
|
|
"num_tokens": 17952809.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 6.08620867729187,
|
|
"epoch": 9.120338140228743,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.032319334768465e-05,
|
|
"loss": 5.0365,
|
|
"mean_token_accuracy": 0.2342731013894081,
|
|
"num_tokens": 17962858.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 6.101832294464112,
|
|
"epoch": 9.125310790651417,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.020658161677821e-05,
|
|
"loss": 5.0638,
|
|
"mean_token_accuracy": 0.2274306520819664,
|
|
"num_tokens": 17971714.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 6.116423273086548,
|
|
"epoch": 9.130283441074093,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.009061697403609e-05,
|
|
"loss": 5.0908,
|
|
"mean_token_accuracy": 0.23057524114847183,
|
|
"num_tokens": 17981512.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 6.0482758522033695,
|
|
"epoch": 9.135256091496768,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.9975299768814204e-05,
|
|
"loss": 5.0458,
|
|
"mean_token_accuracy": 0.22874563187360764,
|
|
"num_tokens": 17991562.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.932010984420776,
|
|
"epoch": 9.140228741919444,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.9860630348518004e-05,
|
|
"loss": 4.9131,
|
|
"mean_token_accuracy": 0.24784310013055802,
|
|
"num_tokens": 18002100.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 6.016932678222656,
|
|
"epoch": 9.145201392342118,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.974660905860135e-05,
|
|
"loss": 4.9866,
|
|
"mean_token_accuracy": 0.2373383954167366,
|
|
"num_tokens": 18011870.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.97070631980896,
|
|
"epoch": 9.150174042764794,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 5.96332362425656e-05,
|
|
"loss": 4.9761,
|
|
"mean_token_accuracy": 0.2361605778336525,
|
|
"num_tokens": 18021683.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 6.043032693862915,
|
|
"epoch": 9.155146693187469,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 5.952051224195852e-05,
|
|
"loss": 5.09,
|
|
"mean_token_accuracy": 0.22675592303276063,
|
|
"num_tokens": 18031430.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 6.114030933380127,
|
|
"epoch": 9.160119343610145,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.940843739637321e-05,
|
|
"loss": 5.0486,
|
|
"mean_token_accuracy": 0.22776918709278107,
|
|
"num_tokens": 18041072.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 6.155849838256836,
|
|
"epoch": 9.165091994032819,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.929701204344712e-05,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.21942081451416015,
|
|
"num_tokens": 18051847.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 6.069553852081299,
|
|
"epoch": 9.170064644455495,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.91862365188611e-05,
|
|
"loss": 5.1302,
|
|
"mean_token_accuracy": 0.22044774144887924,
|
|
"num_tokens": 18061901.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 6.123509311676026,
|
|
"epoch": 9.17503729487817,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.907611115633819e-05,
|
|
"loss": 5.1033,
|
|
"mean_token_accuracy": 0.23151084184646606,
|
|
"num_tokens": 18071492.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 6.168158483505249,
|
|
"epoch": 9.180009945300846,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.896663628764294e-05,
|
|
"loss": 5.1016,
|
|
"mean_token_accuracy": 0.2249367728829384,
|
|
"num_tokens": 18081309.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 6.060184144973755,
|
|
"epoch": 9.18498259572352,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.885781224258002e-05,
|
|
"loss": 5.047,
|
|
"mean_token_accuracy": 0.22480640858411788,
|
|
"num_tokens": 18090970.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 6.039292860031128,
|
|
"epoch": 9.189955246146196,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.8749639348993537e-05,
|
|
"loss": 5.0807,
|
|
"mean_token_accuracy": 0.2235400602221489,
|
|
"num_tokens": 18100858.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 6.123948335647583,
|
|
"epoch": 9.19492789656887,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.864211793276597e-05,
|
|
"loss": 5.1164,
|
|
"mean_token_accuracy": 0.21874127835035323,
|
|
"num_tokens": 18110497.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 6.125820207595825,
|
|
"epoch": 9.199900546991547,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.853524831781702e-05,
|
|
"loss": 5.1012,
|
|
"mean_token_accuracy": 0.22234423756599425,
|
|
"num_tokens": 18119963.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 6.113291072845459,
|
|
"epoch": 9.204873197414221,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.8429030826102886e-05,
|
|
"loss": 5.1073,
|
|
"mean_token_accuracy": 0.2291182905435562,
|
|
"num_tokens": 18129403.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 6.092051219940186,
|
|
"epoch": 9.209845847836897,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.83234657776151e-05,
|
|
"loss": 5.1252,
|
|
"mean_token_accuracy": 0.22523390054702758,
|
|
"num_tokens": 18139917.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 6.040671253204346,
|
|
"epoch": 9.214818498259572,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.8218553490379686e-05,
|
|
"loss": 5.0022,
|
|
"mean_token_accuracy": 0.23213851153850557,
|
|
"num_tokens": 18148781.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.96973066329956,
|
|
"epoch": 9.219791148682248,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 5.8114294280456184e-05,
|
|
"loss": 4.9149,
|
|
"mean_token_accuracy": 0.23867229819297792,
|
|
"num_tokens": 18158856.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 6.047133159637451,
|
|
"epoch": 9.224763799104922,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.801068846193656e-05,
|
|
"loss": 5.0392,
|
|
"mean_token_accuracy": 0.22794160842895508,
|
|
"num_tokens": 18167859.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 6.029630374908447,
|
|
"epoch": 9.229736449527598,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.7907736346944444e-05,
|
|
"loss": 5.0785,
|
|
"mean_token_accuracy": 0.23146802335977554,
|
|
"num_tokens": 18178010.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 6.060396671295166,
|
|
"epoch": 9.234709099950274,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.780543824563411e-05,
|
|
"loss": 5.1022,
|
|
"mean_token_accuracy": 0.22537470906972884,
|
|
"num_tokens": 18188017.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 6.061607360839844,
|
|
"epoch": 9.239681750372949,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.7703794466189524e-05,
|
|
"loss": 5.0325,
|
|
"mean_token_accuracy": 0.22599672228097917,
|
|
"num_tokens": 18197502.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 6.089073944091797,
|
|
"epoch": 9.244654400795625,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.7602805314823495e-05,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.22392918765544892,
|
|
"num_tokens": 18207182.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 6.146832275390625,
|
|
"epoch": 9.2496270512183,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.7502471095776585e-05,
|
|
"loss": 5.1426,
|
|
"mean_token_accuracy": 0.22841822654008864,
|
|
"num_tokens": 18217490.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.952136754989624,
|
|
"epoch": 9.254599701640975,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.740279211131645e-05,
|
|
"loss": 4.9255,
|
|
"mean_token_accuracy": 0.23774073719978334,
|
|
"num_tokens": 18227620.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 6.048126840591431,
|
|
"epoch": 9.25957235206365,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.730376866173662e-05,
|
|
"loss": 5.0117,
|
|
"mean_token_accuracy": 0.22923583686351776,
|
|
"num_tokens": 18237413.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.972932434082031,
|
|
"epoch": 9.264545002486326,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 5.720540104535587e-05,
|
|
"loss": 4.955,
|
|
"mean_token_accuracy": 0.24052845686674118,
|
|
"num_tokens": 18247688.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 6.046155405044556,
|
|
"epoch": 9.269517652909,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.710768955851721e-05,
|
|
"loss": 5.0074,
|
|
"mean_token_accuracy": 0.2383929595351219,
|
|
"num_tokens": 18257248.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 6.075024795532227,
|
|
"epoch": 9.274490303331676,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.701063449558688e-05,
|
|
"loss": 5.0528,
|
|
"mean_token_accuracy": 0.22556237429380416,
|
|
"num_tokens": 18266426.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 6.110498285293579,
|
|
"epoch": 9.27946295375435,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.691423614895373e-05,
|
|
"loss": 5.0595,
|
|
"mean_token_accuracy": 0.23233072459697723,
|
|
"num_tokens": 18275345.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.848920154571533,
|
|
"epoch": 9.284435604177027,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.681849480902806e-05,
|
|
"loss": 4.8164,
|
|
"mean_token_accuracy": 0.265485380589962,
|
|
"num_tokens": 18286156.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 6.114956092834473,
|
|
"epoch": 9.289408254599701,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.672341076424086e-05,
|
|
"loss": 5.1003,
|
|
"mean_token_accuracy": 0.2245888039469719,
|
|
"num_tokens": 18295701.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 6.044072914123535,
|
|
"epoch": 9.294380905022377,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.662898430104306e-05,
|
|
"loss": 5.0609,
|
|
"mean_token_accuracy": 0.22554071992635727,
|
|
"num_tokens": 18305900.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.9980247020721436,
|
|
"epoch": 9.299353555445052,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.653521570390442e-05,
|
|
"loss": 5.0381,
|
|
"mean_token_accuracy": 0.23335582613945008,
|
|
"num_tokens": 18316351.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.996499347686767,
|
|
"epoch": 9.304326205867728,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 5.64421052553129e-05,
|
|
"loss": 4.9865,
|
|
"mean_token_accuracy": 0.24272081404924392,
|
|
"num_tokens": 18326461.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 6.07801456451416,
|
|
"epoch": 9.309298856290402,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.6349653235773774e-05,
|
|
"loss": 5.071,
|
|
"mean_token_accuracy": 0.2367275670170784,
|
|
"num_tokens": 18335792.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 6.099119853973389,
|
|
"epoch": 9.314271506713078,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.625785992380854e-05,
|
|
"loss": 5.1179,
|
|
"mean_token_accuracy": 0.2241745486855507,
|
|
"num_tokens": 18346613.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 6.028288078308106,
|
|
"epoch": 9.319244157135753,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.6166725595954425e-05,
|
|
"loss": 5.0114,
|
|
"mean_token_accuracy": 0.2399899736046791,
|
|
"num_tokens": 18355435.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.993922138214112,
|
|
"epoch": 9.324216807558429,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.6076250526763326e-05,
|
|
"loss": 4.9686,
|
|
"mean_token_accuracy": 0.24012103527784348,
|
|
"num_tokens": 18363518.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 6.139785718917847,
|
|
"epoch": 9.329189457981103,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.598643498880108e-05,
|
|
"loss": 5.1134,
|
|
"mean_token_accuracy": 0.2275055781006813,
|
|
"num_tokens": 18374890.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 6.082642602920532,
|
|
"epoch": 9.33416210840378,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.589727925264665e-05,
|
|
"loss": 5.0487,
|
|
"mean_token_accuracy": 0.22422033101320266,
|
|
"num_tokens": 18384176.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 6.075798225402832,
|
|
"epoch": 9.339134758826454,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.5808783586891186e-05,
|
|
"loss": 5.0393,
|
|
"mean_token_accuracy": 0.23359053432941437,
|
|
"num_tokens": 18393440.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 6.053361606597901,
|
|
"epoch": 9.34410740924913,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.5720948258137345e-05,
|
|
"loss": 4.988,
|
|
"mean_token_accuracy": 0.244154717028141,
|
|
"num_tokens": 18403058.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 6.039453840255737,
|
|
"epoch": 9.349080059671804,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 5.563377353099849e-05,
|
|
"loss": 5.0558,
|
|
"mean_token_accuracy": 0.22714245319366455,
|
|
"num_tokens": 18414189.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 6.0394734859466555,
|
|
"epoch": 9.35405271009448,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.55472596680978e-05,
|
|
"loss": 5.0403,
|
|
"mean_token_accuracy": 0.23600122183561326,
|
|
"num_tokens": 18423438.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.991438865661621,
|
|
"epoch": 9.359025360517156,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.546140693006755e-05,
|
|
"loss": 5.0199,
|
|
"mean_token_accuracy": 0.23588869571685792,
|
|
"num_tokens": 18433060.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 6.064379453659058,
|
|
"epoch": 9.36399801093983,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.537621557554831e-05,
|
|
"loss": 5.0557,
|
|
"mean_token_accuracy": 0.2291693225502968,
|
|
"num_tokens": 18442687.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 6.080716705322265,
|
|
"epoch": 9.368970661362507,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.529168586118814e-05,
|
|
"loss": 5.1085,
|
|
"mean_token_accuracy": 0.21872063875198364,
|
|
"num_tokens": 18451990.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 6.094289779663086,
|
|
"epoch": 9.373943311785181,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 5.5207818041641894e-05,
|
|
"loss": 5.0967,
|
|
"mean_token_accuracy": 0.23390048295259475,
|
|
"num_tokens": 18462826.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 6.087824010848999,
|
|
"epoch": 9.378915962207858,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.5124612369570304e-05,
|
|
"loss": 5.0807,
|
|
"mean_token_accuracy": 0.22836389541625976,
|
|
"num_tokens": 18472466.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 6.130790281295776,
|
|
"epoch": 9.383888612630532,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.504206909563937e-05,
|
|
"loss": 5.2072,
|
|
"mean_token_accuracy": 0.20930440723896027,
|
|
"num_tokens": 18483008.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 6.148341941833496,
|
|
"epoch": 9.388861263053208,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.496018846851957e-05,
|
|
"loss": 5.1784,
|
|
"mean_token_accuracy": 0.2142087310552597,
|
|
"num_tokens": 18493321.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 6.041688346862793,
|
|
"epoch": 9.393833913475882,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.4878970734885066e-05,
|
|
"loss": 4.9832,
|
|
"mean_token_accuracy": 0.24046161621809006,
|
|
"num_tokens": 18502438.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 6.123617792129517,
|
|
"epoch": 9.398806563898559,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 5.479841613941291e-05,
|
|
"loss": 5.1573,
|
|
"mean_token_accuracy": 0.22786664217710495,
|
|
"num_tokens": 18512839.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 6.0788451671600345,
|
|
"epoch": 9.403779214321233,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.471852492478251e-05,
|
|
"loss": 5.0493,
|
|
"mean_token_accuracy": 0.23044502586126328,
|
|
"num_tokens": 18522431.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.9342334270477295,
|
|
"epoch": 9.408751864743909,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.4639297331674625e-05,
|
|
"loss": 4.903,
|
|
"mean_token_accuracy": 0.24853922575712203,
|
|
"num_tokens": 18532889.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 6.066343927383423,
|
|
"epoch": 9.413724515166583,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.4560733598770956e-05,
|
|
"loss": 5.0547,
|
|
"mean_token_accuracy": 0.2258471205830574,
|
|
"num_tokens": 18541937.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.715981435775757,
|
|
"epoch": 9.41869716558926,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.4482833962753094e-05,
|
|
"loss": 4.5882,
|
|
"mean_token_accuracy": 0.2845833048224449,
|
|
"num_tokens": 18552107.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 6.132922220230102,
|
|
"epoch": 9.423669816011934,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.440559865830207e-05,
|
|
"loss": 5.1907,
|
|
"mean_token_accuracy": 0.21502602100372314,
|
|
"num_tokens": 18562075.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 6.102835988998413,
|
|
"epoch": 9.42864246643461,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 5.432902791809747e-05,
|
|
"loss": 5.0551,
|
|
"mean_token_accuracy": 0.22030193954706193,
|
|
"num_tokens": 18572325.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.974186515808105,
|
|
"epoch": 9.433615116857284,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.425312197281686e-05,
|
|
"loss": 4.927,
|
|
"mean_token_accuracy": 0.24597188085317612,
|
|
"num_tokens": 18582926.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 6.00345664024353,
|
|
"epoch": 9.43858776727996,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.417788105113501e-05,
|
|
"loss": 4.9521,
|
|
"mean_token_accuracy": 0.24089189767837524,
|
|
"num_tokens": 18592869.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.9997974872589115,
|
|
"epoch": 9.443560417702635,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.4103305379723265e-05,
|
|
"loss": 4.9681,
|
|
"mean_token_accuracy": 0.24150547981262208,
|
|
"num_tokens": 18602337.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 6.1114105701446535,
|
|
"epoch": 9.448533068125311,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.4029395183248844e-05,
|
|
"loss": 5.1473,
|
|
"mean_token_accuracy": 0.22289662510156633,
|
|
"num_tokens": 18612393.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 6.076411008834839,
|
|
"epoch": 9.453505718547985,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.3956150684374093e-05,
|
|
"loss": 5.0763,
|
|
"mean_token_accuracy": 0.2246367633342743,
|
|
"num_tokens": 18621701.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 6.075271129608154,
|
|
"epoch": 9.458478368970662,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.388357210375588e-05,
|
|
"loss": 5.0127,
|
|
"mean_token_accuracy": 0.23302044570446015,
|
|
"num_tokens": 18631584.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 6.083095598220825,
|
|
"epoch": 9.463451019393336,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.381165966004503e-05,
|
|
"loss": 5.0142,
|
|
"mean_token_accuracy": 0.24064901769161223,
|
|
"num_tokens": 18640881.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 6.134675121307373,
|
|
"epoch": 9.468423669816012,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.3740413569885366e-05,
|
|
"loss": 5.1352,
|
|
"mean_token_accuracy": 0.22577565610408784,
|
|
"num_tokens": 18649731.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 6.113027620315552,
|
|
"epoch": 9.473396320238686,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 5.366983404791347e-05,
|
|
"loss": 5.1674,
|
|
"mean_token_accuracy": 0.22000565528869628,
|
|
"num_tokens": 18661797.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 6.043666124343872,
|
|
"epoch": 9.478368970661363,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.359992130675761e-05,
|
|
"loss": 5.0532,
|
|
"mean_token_accuracy": 0.23343214243650437,
|
|
"num_tokens": 18671766.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 6.043400669097901,
|
|
"epoch": 9.483341621084037,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.3530675557037486e-05,
|
|
"loss": 4.9224,
|
|
"mean_token_accuracy": 0.24238812923431396,
|
|
"num_tokens": 18681757.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 6.069650030136108,
|
|
"epoch": 9.488314271506713,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.3462097007363274e-05,
|
|
"loss": 4.9936,
|
|
"mean_token_accuracy": 0.24252108186483384,
|
|
"num_tokens": 18691218.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 6.092237424850464,
|
|
"epoch": 9.49328692192939,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.3394185864335225e-05,
|
|
"loss": 5.0654,
|
|
"mean_token_accuracy": 0.23353236317634582,
|
|
"num_tokens": 18701006.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 6.09539384841919,
|
|
"epoch": 9.498259572352064,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.3326942332542907e-05,
|
|
"loss": 5.0658,
|
|
"mean_token_accuracy": 0.22592489868402482,
|
|
"num_tokens": 18710790.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 6.060313749313354,
|
|
"epoch": 9.50323222277474,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.326036661456468e-05,
|
|
"loss": 5.0519,
|
|
"mean_token_accuracy": 0.23020283579826356,
|
|
"num_tokens": 18719609.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 6.066486358642578,
|
|
"epoch": 9.508204873197414,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.319445891096703e-05,
|
|
"loss": 5.0528,
|
|
"mean_token_accuracy": 0.2329959586262703,
|
|
"num_tokens": 18728243.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 6.003104782104492,
|
|
"epoch": 9.51317752362009,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.312921942030392e-05,
|
|
"loss": 4.9369,
|
|
"mean_token_accuracy": 0.24376342296600342,
|
|
"num_tokens": 18737958.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 6.119411849975586,
|
|
"epoch": 9.518150174042765,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.306464833911631e-05,
|
|
"loss": 5.1908,
|
|
"mean_token_accuracy": 0.21767373979091645,
|
|
"num_tokens": 18747758.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 6.087343168258667,
|
|
"epoch": 9.52312282446544,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.300074586193152e-05,
|
|
"loss": 5.0394,
|
|
"mean_token_accuracy": 0.23996152579784394,
|
|
"num_tokens": 18759293.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 6.109319448471069,
|
|
"epoch": 9.528095474888115,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.293751218126261e-05,
|
|
"loss": 5.0277,
|
|
"mean_token_accuracy": 0.23840824365615845,
|
|
"num_tokens": 18769977.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.731642436981201,
|
|
"epoch": 9.533068125310791,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.287494748760776e-05,
|
|
"loss": 4.7146,
|
|
"mean_token_accuracy": 0.27471356391906737,
|
|
"num_tokens": 18780294.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 6.104980611801148,
|
|
"epoch": 9.538040775733466,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.2813051969449886e-05,
|
|
"loss": 5.0374,
|
|
"mean_token_accuracy": 0.23377232998609543,
|
|
"num_tokens": 18789936.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 6.076588726043701,
|
|
"epoch": 9.543013426156142,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.2751825813255794e-05,
|
|
"loss": 5.0474,
|
|
"mean_token_accuracy": 0.23537371456623077,
|
|
"num_tokens": 18799446.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.963018846511841,
|
|
"epoch": 9.547986076578816,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.269126920347583e-05,
|
|
"loss": 4.9472,
|
|
"mean_token_accuracy": 0.23589526414871215,
|
|
"num_tokens": 18809648.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 6.01171646118164,
|
|
"epoch": 9.552958727001492,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.263138232254333e-05,
|
|
"loss": 4.9664,
|
|
"mean_token_accuracy": 0.2387266919016838,
|
|
"num_tokens": 18820334.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 6.156089353561401,
|
|
"epoch": 9.557931377424167,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.257216535087386e-05,
|
|
"loss": 5.0964,
|
|
"mean_token_accuracy": 0.2261794239282608,
|
|
"num_tokens": 18829715.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 6.091220474243164,
|
|
"epoch": 9.562904027846843,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.251361846686495e-05,
|
|
"loss": 5.0594,
|
|
"mean_token_accuracy": 0.23306489884853362,
|
|
"num_tokens": 18839602.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 6.076358461380005,
|
|
"epoch": 9.567876678269517,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.245574184689531e-05,
|
|
"loss": 5.0957,
|
|
"mean_token_accuracy": 0.23093206733465194,
|
|
"num_tokens": 18849441.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 6.093290996551514,
|
|
"epoch": 9.572849328692193,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.239853566532444e-05,
|
|
"loss": 5.1096,
|
|
"mean_token_accuracy": 0.22312760204076768,
|
|
"num_tokens": 18859695.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 6.026937246322632,
|
|
"epoch": 9.577821979114868,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.234200009449215e-05,
|
|
"loss": 4.9823,
|
|
"mean_token_accuracy": 0.2426535204052925,
|
|
"num_tokens": 18868660.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 6.056035995483398,
|
|
"epoch": 9.582794629537544,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.228613530471782e-05,
|
|
"loss": 5.045,
|
|
"mean_token_accuracy": 0.23108498454093934,
|
|
"num_tokens": 18878101.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 6.112252950668335,
|
|
"epoch": 9.587767279960218,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.223094146430017e-05,
|
|
"loss": 5.1709,
|
|
"mean_token_accuracy": 0.22449235171079635,
|
|
"num_tokens": 18887202.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 6.07296233177185,
|
|
"epoch": 9.592739930382894,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.2176418739516534e-05,
|
|
"loss": 5.1293,
|
|
"mean_token_accuracy": 0.2290360301733017,
|
|
"num_tokens": 18897958.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 6.148199224472046,
|
|
"epoch": 9.597712580805569,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 5.212256729462242e-05,
|
|
"loss": 5.1725,
|
|
"mean_token_accuracy": 0.21796624809503556,
|
|
"num_tokens": 18907510.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.98064112663269,
|
|
"epoch": 9.602685231228245,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 5.2069387291851136e-05,
|
|
"loss": 4.9261,
|
|
"mean_token_accuracy": 0.2445056989789009,
|
|
"num_tokens": 18919219.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 6.0106054782867435,
|
|
"epoch": 9.607657881650919,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 5.2016878891413096e-05,
|
|
"loss": 4.9185,
|
|
"mean_token_accuracy": 0.2439217045903206,
|
|
"num_tokens": 18928993.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.9511988162994385,
|
|
"epoch": 9.612630532073595,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 5.196504225149547e-05,
|
|
"loss": 5.0343,
|
|
"mean_token_accuracy": 0.23529305011034013,
|
|
"num_tokens": 18939154.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 6.088329315185547,
|
|
"epoch": 9.617603182496271,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.19138775282617e-05,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.22451289296150206,
|
|
"num_tokens": 18948265.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 6.063173007965088,
|
|
"epoch": 9.622575832918946,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 5.186338487585098e-05,
|
|
"loss": 5.0824,
|
|
"mean_token_accuracy": 0.22483629137277603,
|
|
"num_tokens": 18957390.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 6.082346963882446,
|
|
"epoch": 9.627548483341622,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.1813564446377855e-05,
|
|
"loss": 5.037,
|
|
"mean_token_accuracy": 0.23511048257350922,
|
|
"num_tokens": 18966491.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 6.104487037658691,
|
|
"epoch": 9.632521133764296,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.176441638993165e-05,
|
|
"loss": 5.1646,
|
|
"mean_token_accuracy": 0.22917530536651612,
|
|
"num_tokens": 18976450.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 6.043255758285523,
|
|
"epoch": 9.637493784186972,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.1715940854576196e-05,
|
|
"loss": 5.0368,
|
|
"mean_token_accuracy": 0.232001955807209,
|
|
"num_tokens": 18985749.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 6.080848741531372,
|
|
"epoch": 9.642466434609647,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.1668137986349215e-05,
|
|
"loss": 5.1229,
|
|
"mean_token_accuracy": 0.22467821687459946,
|
|
"num_tokens": 18996288.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 6.115914058685303,
|
|
"epoch": 9.647439085032323,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.1621007929261955e-05,
|
|
"loss": 5.0805,
|
|
"mean_token_accuracy": 0.2311658129096031,
|
|
"num_tokens": 19006340.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 6.049308490753174,
|
|
"epoch": 9.652411735454997,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.157455082529877e-05,
|
|
"loss": 5.0036,
|
|
"mean_token_accuracy": 0.231049245595932,
|
|
"num_tokens": 19015773.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 6.069045066833496,
|
|
"epoch": 9.657384385877673,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.1528766814416664e-05,
|
|
"loss": 5.0334,
|
|
"mean_token_accuracy": 0.23660432249307634,
|
|
"num_tokens": 19024547.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 6.108839416503907,
|
|
"epoch": 9.662357036300348,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.1483656034544846e-05,
|
|
"loss": 5.1021,
|
|
"mean_token_accuracy": 0.2262729972600937,
|
|
"num_tokens": 19034432.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 6.109891176223755,
|
|
"epoch": 9.667329686723024,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.14392186215844e-05,
|
|
"loss": 5.0495,
|
|
"mean_token_accuracy": 0.23054512739181518,
|
|
"num_tokens": 19044032.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 6.092459440231323,
|
|
"epoch": 9.672302337145698,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 5.1395454709407746e-05,
|
|
"loss": 5.0805,
|
|
"mean_token_accuracy": 0.223032084107399,
|
|
"num_tokens": 19054706.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 6.031870317459107,
|
|
"epoch": 9.677274987568374,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.1352364429858414e-05,
|
|
"loss": 4.9267,
|
|
"mean_token_accuracy": 0.24563615769147873,
|
|
"num_tokens": 19063280.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 6.110308504104614,
|
|
"epoch": 9.682247637991049,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.1309947912750456e-05,
|
|
"loss": 5.1466,
|
|
"mean_token_accuracy": 0.21846730560064315,
|
|
"num_tokens": 19073006.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 6.043213272094727,
|
|
"epoch": 9.687220288413725,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.126820528586815e-05,
|
|
"loss": 5.0961,
|
|
"mean_token_accuracy": 0.2257940724492073,
|
|
"num_tokens": 19082797.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 6.040573692321777,
|
|
"epoch": 9.6921929388364,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.1227136674965646e-05,
|
|
"loss": 4.9968,
|
|
"mean_token_accuracy": 0.23239275813102722,
|
|
"num_tokens": 19092126.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 6.012143325805664,
|
|
"epoch": 9.697165589259075,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.11867422037665e-05,
|
|
"loss": 5.0873,
|
|
"mean_token_accuracy": 0.22992108166217803,
|
|
"num_tokens": 19102662.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.878934907913208,
|
|
"epoch": 9.70213823968175,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.114702199396343e-05,
|
|
"loss": 4.8437,
|
|
"mean_token_accuracy": 0.252801014482975,
|
|
"num_tokens": 19112430.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 6.033525085449218,
|
|
"epoch": 9.707110890104426,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 5.1107976165217795e-05,
|
|
"loss": 5.0344,
|
|
"mean_token_accuracy": 0.22860213220119477,
|
|
"num_tokens": 19122540.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 6.03737096786499,
|
|
"epoch": 9.7120835405271,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.106960483515934e-05,
|
|
"loss": 5.0534,
|
|
"mean_token_accuracy": 0.22924555838108063,
|
|
"num_tokens": 19131636.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 6.139129590988159,
|
|
"epoch": 9.717056190949776,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.10319081193858e-05,
|
|
"loss": 5.1426,
|
|
"mean_token_accuracy": 0.21775369346141815,
|
|
"num_tokens": 19141865.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 6.096522521972656,
|
|
"epoch": 9.72202884137245,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.099488613146255e-05,
|
|
"loss": 5.132,
|
|
"mean_token_accuracy": 0.22761220782995223,
|
|
"num_tokens": 19152075.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.942059421539307,
|
|
"epoch": 9.727001491795127,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.095853898292233e-05,
|
|
"loss": 4.9023,
|
|
"mean_token_accuracy": 0.25206171721220016,
|
|
"num_tokens": 19162503.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 6.042265319824219,
|
|
"epoch": 9.731974142217801,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.0922866783264784e-05,
|
|
"loss": 5.0519,
|
|
"mean_token_accuracy": 0.2344980776309967,
|
|
"num_tokens": 19172317.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 6.0998128890991214,
|
|
"epoch": 9.736946792640477,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.088786963995624e-05,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.22139425128698348,
|
|
"num_tokens": 19181388.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 6.133503866195679,
|
|
"epoch": 9.741919443063153,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.0853547658429324e-05,
|
|
"loss": 5.0912,
|
|
"mean_token_accuracy": 0.22566074281930923,
|
|
"num_tokens": 19191625.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 6.12084927558899,
|
|
"epoch": 9.746892093485828,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.081990094208271e-05,
|
|
"loss": 5.055,
|
|
"mean_token_accuracy": 0.22512171864509584,
|
|
"num_tokens": 19200359.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.937306213378906,
|
|
"epoch": 9.751864743908504,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 5.0786929592280706e-05,
|
|
"loss": 4.9574,
|
|
"mean_token_accuracy": 0.24423912018537522,
|
|
"num_tokens": 19210514.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 6.1132646083831785,
|
|
"epoch": 9.756837394331178,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 5.075463370835307e-05,
|
|
"loss": 5.1769,
|
|
"mean_token_accuracy": 0.2167748063802719,
|
|
"num_tokens": 19222365.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 6.109905004501343,
|
|
"epoch": 9.761810044753854,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.072301338759455e-05,
|
|
"loss": 5.1208,
|
|
"mean_token_accuracy": 0.2238441824913025,
|
|
"num_tokens": 19231902.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 6.119198369979858,
|
|
"epoch": 9.766782695176529,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.069206872526479e-05,
|
|
"loss": 5.0646,
|
|
"mean_token_accuracy": 0.22557778805494308,
|
|
"num_tokens": 19241260.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 6.033057308197021,
|
|
"epoch": 9.771755345599205,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 5.06617998145879e-05,
|
|
"loss": 4.9345,
|
|
"mean_token_accuracy": 0.24134642630815506,
|
|
"num_tokens": 19250500.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 6.02874608039856,
|
|
"epoch": 9.77672799602188,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.063220674675219e-05,
|
|
"loss": 4.9516,
|
|
"mean_token_accuracy": 0.24385342597961426,
|
|
"num_tokens": 19260272.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 6.122976016998291,
|
|
"epoch": 9.781700646444556,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.060328961090996e-05,
|
|
"loss": 5.0915,
|
|
"mean_token_accuracy": 0.21696975082159042,
|
|
"num_tokens": 19269807.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 6.166137552261352,
|
|
"epoch": 9.78667329686723,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.057504849417717e-05,
|
|
"loss": 5.1016,
|
|
"mean_token_accuracy": 0.22387044727802277,
|
|
"num_tokens": 19279758.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 6.094998168945312,
|
|
"epoch": 9.791645947289906,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.05474834816332e-05,
|
|
"loss": 5.0673,
|
|
"mean_token_accuracy": 0.22979679852724075,
|
|
"num_tokens": 19290936.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.997396802902221,
|
|
"epoch": 9.79661859771258,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.052059465632063e-05,
|
|
"loss": 5.0149,
|
|
"mean_token_accuracy": 0.23517657220363616,
|
|
"num_tokens": 19299448.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 6.110661602020263,
|
|
"epoch": 9.801591248135257,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.049438209924489e-05,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.2232228100299835,
|
|
"num_tokens": 19309926.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 6.13741455078125,
|
|
"epoch": 9.80656389855793,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 5.0468845889374115e-05,
|
|
"loss": 5.1405,
|
|
"mean_token_accuracy": 0.22344421595335007,
|
|
"num_tokens": 19320495.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.891877841949463,
|
|
"epoch": 9.811536548980607,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 5.04439861036389e-05,
|
|
"loss": 4.7115,
|
|
"mean_token_accuracy": 0.26418369114398954,
|
|
"num_tokens": 19329699.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 6.098142385482788,
|
|
"epoch": 9.816509199403281,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.041980281693198e-05,
|
|
"loss": 4.9808,
|
|
"mean_token_accuracy": 0.23785579949617386,
|
|
"num_tokens": 19338760.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 6.019230842590332,
|
|
"epoch": 9.821481849825958,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.0396296102108134e-05,
|
|
"loss": 4.9866,
|
|
"mean_token_accuracy": 0.23784855157136917,
|
|
"num_tokens": 19349098.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 6.025642967224121,
|
|
"epoch": 9.826454500248632,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.037346602998383e-05,
|
|
"loss": 5.0758,
|
|
"mean_token_accuracy": 0.2301526516675949,
|
|
"num_tokens": 19359541.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 6.043281984329224,
|
|
"epoch": 9.831427150671308,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.03513126693371e-05,
|
|
"loss": 4.9985,
|
|
"mean_token_accuracy": 0.2363782048225403,
|
|
"num_tokens": 19369376.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 6.078520822525024,
|
|
"epoch": 9.836399801093982,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 5.0329836086907294e-05,
|
|
"loss": 5.0097,
|
|
"mean_token_accuracy": 0.2343876913189888,
|
|
"num_tokens": 19379147.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 6.0680419445037845,
|
|
"epoch": 9.841372451516659,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.030903634739498e-05,
|
|
"loss": 5.0799,
|
|
"mean_token_accuracy": 0.2251813605427742,
|
|
"num_tokens": 19388714.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 6.062436628341675,
|
|
"epoch": 9.846345101939333,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.028891351346154e-05,
|
|
"loss": 5.0225,
|
|
"mean_token_accuracy": 0.23517635017633437,
|
|
"num_tokens": 19397495.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 6.075806283950806,
|
|
"epoch": 9.851317752362009,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 5.0269467645729184e-05,
|
|
"loss": 5.0375,
|
|
"mean_token_accuracy": 0.22746315747499465,
|
|
"num_tokens": 19407369.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.935221242904663,
|
|
"epoch": 9.856290402784683,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.0250698802780684e-05,
|
|
"loss": 4.9274,
|
|
"mean_token_accuracy": 0.2507831111550331,
|
|
"num_tokens": 19417169.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 6.134442663192749,
|
|
"epoch": 9.86126305320736,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.0232607041159164e-05,
|
|
"loss": 5.0978,
|
|
"mean_token_accuracy": 0.2177930384874344,
|
|
"num_tokens": 19426463.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 6.064385271072387,
|
|
"epoch": 9.866235703630036,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.0215192415368e-05,
|
|
"loss": 4.9972,
|
|
"mean_token_accuracy": 0.23357438892126084,
|
|
"num_tokens": 19436938.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 6.1256482124328615,
|
|
"epoch": 9.87120835405271,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 5.0198454977870655e-05,
|
|
"loss": 5.1194,
|
|
"mean_token_accuracy": 0.22843537479639053,
|
|
"num_tokens": 19446930.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 6.130322980880737,
|
|
"epoch": 9.876181004475386,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.018239477909039e-05,
|
|
"loss": 5.1595,
|
|
"mean_token_accuracy": 0.2192773178219795,
|
|
"num_tokens": 19456344.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 6.020367383956909,
|
|
"epoch": 9.88115365489806,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.01670118674103e-05,
|
|
"loss": 5.0258,
|
|
"mean_token_accuracy": 0.23473658561706542,
|
|
"num_tokens": 19466553.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 6.085891151428223,
|
|
"epoch": 9.886126305320737,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.015230628917305e-05,
|
|
"loss": 5.1001,
|
|
"mean_token_accuracy": 0.22515126019716264,
|
|
"num_tokens": 19476567.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 6.156118774414063,
|
|
"epoch": 9.891098955743411,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.013827808868079e-05,
|
|
"loss": 5.1656,
|
|
"mean_token_accuracy": 0.22421378940343856,
|
|
"num_tokens": 19485670.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.924436950683594,
|
|
"epoch": 9.896071606166087,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.012492730819495e-05,
|
|
"loss": 4.8989,
|
|
"mean_token_accuracy": 0.24664329141378402,
|
|
"num_tokens": 19496315.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 6.08775749206543,
|
|
"epoch": 9.901044256588762,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 5.0112253987936196e-05,
|
|
"loss": 5.0715,
|
|
"mean_token_accuracy": 0.22801166623830796,
|
|
"num_tokens": 19505932.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 6.101383876800537,
|
|
"epoch": 9.906016907011438,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.0100258166084246e-05,
|
|
"loss": 5.1327,
|
|
"mean_token_accuracy": 0.21574556976556777,
|
|
"num_tokens": 19516912.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 6.084262180328369,
|
|
"epoch": 9.910989557434112,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 5.008893987877781e-05,
|
|
"loss": 5.0786,
|
|
"mean_token_accuracy": 0.22036135345697402,
|
|
"num_tokens": 19526095.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.957286691665649,
|
|
"epoch": 9.915962207856788,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.0078299160114443e-05,
|
|
"loss": 4.9495,
|
|
"mean_token_accuracy": 0.24328897446393966,
|
|
"num_tokens": 19537147.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 6.105463171005249,
|
|
"epoch": 9.920934858279463,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 5.0068336042150455e-05,
|
|
"loss": 5.0356,
|
|
"mean_token_accuracy": 0.22863925248384476,
|
|
"num_tokens": 19546014.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 6.139309120178223,
|
|
"epoch": 9.925907508702139,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.005905055490081e-05,
|
|
"loss": 5.1597,
|
|
"mean_token_accuracy": 0.21593999713659287,
|
|
"num_tokens": 19555451.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 6.086447525024414,
|
|
"epoch": 9.930880159124813,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.0050442726339e-05,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.23960970640182494,
|
|
"num_tokens": 19563773.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.76749358177185,
|
|
"epoch": 9.93585280954749,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.004251258239703e-05,
|
|
"loss": 4.7344,
|
|
"mean_token_accuracy": 0.27450508922338485,
|
|
"num_tokens": 19573821.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.9696718692779545,
|
|
"epoch": 9.940825459970164,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 5.0035260146965355e-05,
|
|
"loss": 4.9115,
|
|
"mean_token_accuracy": 0.24854292273521422,
|
|
"num_tokens": 19584768.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 9.940825459970164,
|
|
"eval_entropy": 5.9434962481397635,
|
|
"eval_loss": 6.453098297119141,
|
|
"eval_mean_token_accuracy": 0.1508507455445929,
|
|
"eval_num_tokens": 19584768.0,
|
|
"eval_runtime": 2.1336,
|
|
"eval_samples_per_second": 1624.019,
|
|
"eval_steps_per_second": 203.412,
|
|
"step": 10000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10050,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4759366380503040.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|