2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.9945300845350572,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 477704557854720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|