4057 lines
109 KiB
JSON
4057 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.9885629040278467,
|
|
"eval_steps": 1000,
|
|
"global_step": 2000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 956740866508800.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|