4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.7749667110519307,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692031192779542,
|
|
"epoch": 0.004438526409232135,
|
|
"grad_norm": 14.75,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8207,
|
|
"mean_token_accuracy": 0.000128369708545506,
|
|
"num_tokens": 9810.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691952133178711,
|
|
"epoch": 0.00887705281846427,
|
|
"grad_norm": 12.9375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7613,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20538.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690592384338379,
|
|
"epoch": 0.013315579227696404,
|
|
"grad_norm": 9.9375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5054,
|
|
"mean_token_accuracy": 0.03474730136804283,
|
|
"num_tokens": 30258.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.66368646621704,
|
|
"epoch": 0.01775410563692854,
|
|
"grad_norm": 5.875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1947,
|
|
"mean_token_accuracy": 0.04976645708084106,
|
|
"num_tokens": 41249.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.518089294433594,
|
|
"epoch": 0.022192632046160676,
|
|
"grad_norm": 3.609375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9,
|
|
"mean_token_accuracy": 0.053335465490818024,
|
|
"num_tokens": 51738.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.489847755432129,
|
|
"epoch": 0.02663115845539281,
|
|
"grad_norm": 3.0625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8161,
|
|
"mean_token_accuracy": 0.05364362597465515,
|
|
"num_tokens": 62920.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.506165885925293,
|
|
"epoch": 0.031069684864624945,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7103,
|
|
"mean_token_accuracy": 0.05455525256693363,
|
|
"num_tokens": 73163.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.478147506713867,
|
|
"epoch": 0.03550821127385708,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.682,
|
|
"mean_token_accuracy": 0.05231944099068642,
|
|
"num_tokens": 84180.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.461441326141358,
|
|
"epoch": 0.03994673768308921,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6388,
|
|
"mean_token_accuracy": 0.05221981666982174,
|
|
"num_tokens": 94108.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.476018333435059,
|
|
"epoch": 0.04438526409232135,
|
|
"grad_norm": 3.140625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6253,
|
|
"mean_token_accuracy": 0.05243444815278053,
|
|
"num_tokens": 104696.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.478155612945557,
|
|
"epoch": 0.048823790501553485,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5235,
|
|
"mean_token_accuracy": 0.0584161002188921,
|
|
"num_tokens": 114568.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.39725103378296,
|
|
"epoch": 0.05326231691078562,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4713,
|
|
"mean_token_accuracy": 0.064088149741292,
|
|
"num_tokens": 126251.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.248920154571532,
|
|
"epoch": 0.05770084332001776,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3504,
|
|
"mean_token_accuracy": 0.06839247606694698,
|
|
"num_tokens": 135612.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.320736598968505,
|
|
"epoch": 0.06213936972924989,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4004,
|
|
"mean_token_accuracy": 0.0615400142967701,
|
|
"num_tokens": 145962.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.323652744293213,
|
|
"epoch": 0.06657789613848203,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1911,
|
|
"mean_token_accuracy": 0.07103513963520527,
|
|
"num_tokens": 155587.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.258409976959229,
|
|
"epoch": 0.07101642254771416,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1343,
|
|
"mean_token_accuracy": 0.06803948283195496,
|
|
"num_tokens": 165657.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.24950189590454,
|
|
"epoch": 0.0754549489569463,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1277,
|
|
"mean_token_accuracy": 0.06729609742760659,
|
|
"num_tokens": 175844.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.224775981903075,
|
|
"epoch": 0.07989347536617843,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9736,
|
|
"mean_token_accuracy": 0.06970795653760434,
|
|
"num_tokens": 185453.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.128818893432618,
|
|
"epoch": 0.08433200177541056,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9469,
|
|
"mean_token_accuracy": 0.07015660107135772,
|
|
"num_tokens": 195296.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.158192539215088,
|
|
"epoch": 0.0887705281846427,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8594,
|
|
"mean_token_accuracy": 0.0717735793441534,
|
|
"num_tokens": 204491.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.063592338562012,
|
|
"epoch": 0.09320905459387484,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.8275,
|
|
"mean_token_accuracy": 0.06594390086829663,
|
|
"num_tokens": 214916.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.0551118850708,
|
|
"epoch": 0.09764758100310697,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.6661,
|
|
"mean_token_accuracy": 0.07398808561265469,
|
|
"num_tokens": 226660.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.929056739807129,
|
|
"epoch": 0.1020861074123391,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.6191,
|
|
"mean_token_accuracy": 0.06771207004785537,
|
|
"num_tokens": 236554.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.87773323059082,
|
|
"epoch": 0.10652463382157124,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.5466,
|
|
"mean_token_accuracy": 0.07039828561246395,
|
|
"num_tokens": 247388.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.78554401397705,
|
|
"epoch": 0.11096316023080337,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.3693,
|
|
"mean_token_accuracy": 0.07505594864487648,
|
|
"num_tokens": 257404.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.569751739501953,
|
|
"epoch": 0.11540168664003551,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.366,
|
|
"mean_token_accuracy": 0.0721758782863617,
|
|
"num_tokens": 268024.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.49385757446289,
|
|
"epoch": 0.11984021304926765,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.3116,
|
|
"mean_token_accuracy": 0.07231119871139527,
|
|
"num_tokens": 278477.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.405902671813966,
|
|
"epoch": 0.12427873945849978,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.2427,
|
|
"mean_token_accuracy": 0.06903640553355217,
|
|
"num_tokens": 289974.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.16763277053833,
|
|
"epoch": 0.1287172658677319,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.1839,
|
|
"mean_token_accuracy": 0.06725569888949394,
|
|
"num_tokens": 300150.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.045907497406006,
|
|
"epoch": 0.13315579227696406,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.0648,
|
|
"mean_token_accuracy": 0.07113164104521275,
|
|
"num_tokens": 310482.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.951902961730957,
|
|
"epoch": 0.13759431868619618,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1043,
|
|
"mean_token_accuracy": 0.07190654650330544,
|
|
"num_tokens": 321350.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.734932899475098,
|
|
"epoch": 0.14203284509542832,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9902,
|
|
"mean_token_accuracy": 0.07260116301476956,
|
|
"num_tokens": 332009.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.618322277069092,
|
|
"epoch": 0.14647137150466044,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.0046,
|
|
"mean_token_accuracy": 0.07010005228221416,
|
|
"num_tokens": 341980.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.576034545898438,
|
|
"epoch": 0.1509098979138926,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.9965,
|
|
"mean_token_accuracy": 0.06660077758133412,
|
|
"num_tokens": 354249.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.524139022827148,
|
|
"epoch": 0.15534842432312473,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.9076,
|
|
"mean_token_accuracy": 0.07078476324677467,
|
|
"num_tokens": 364295.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.328704929351806,
|
|
"epoch": 0.15978695073235685,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.8464,
|
|
"mean_token_accuracy": 0.0794072538614273,
|
|
"num_tokens": 373601.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.382467269897461,
|
|
"epoch": 0.164225477141589,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8583,
|
|
"mean_token_accuracy": 0.07436131834983825,
|
|
"num_tokens": 384260.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.316840362548827,
|
|
"epoch": 0.16866400355082112,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.9058,
|
|
"mean_token_accuracy": 0.06711366176605224,
|
|
"num_tokens": 394964.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.204895782470704,
|
|
"epoch": 0.17310252996005326,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.8384,
|
|
"mean_token_accuracy": 0.07343141734600067,
|
|
"num_tokens": 404470.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.270458793640136,
|
|
"epoch": 0.1775410563692854,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.9041,
|
|
"mean_token_accuracy": 0.07145134881138801,
|
|
"num_tokens": 414822.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.344293403625489,
|
|
"epoch": 0.18197958277851753,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.8228,
|
|
"mean_token_accuracy": 0.07206394150853157,
|
|
"num_tokens": 424367.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.169469261169434,
|
|
"epoch": 0.18641810918774968,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.8063,
|
|
"mean_token_accuracy": 0.06796807795763016,
|
|
"num_tokens": 434911.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.127049207687378,
|
|
"epoch": 0.1908566355969818,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.7417,
|
|
"mean_token_accuracy": 0.08103418126702308,
|
|
"num_tokens": 445215.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.216034698486329,
|
|
"epoch": 0.19529516200621394,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.7919,
|
|
"mean_token_accuracy": 0.07842708677053452,
|
|
"num_tokens": 454786.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.140088844299317,
|
|
"epoch": 0.19973368841544606,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.8219,
|
|
"mean_token_accuracy": 0.07302889823913575,
|
|
"num_tokens": 465185.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.021256399154662,
|
|
"epoch": 0.2041722148246782,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.7156,
|
|
"mean_token_accuracy": 0.08027704544365406,
|
|
"num_tokens": 475235.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.17042875289917,
|
|
"epoch": 0.20861074123391035,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7317,
|
|
"mean_token_accuracy": 0.07805200070142745,
|
|
"num_tokens": 485408.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.040202140808105,
|
|
"epoch": 0.21304926764314247,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.7331,
|
|
"mean_token_accuracy": 0.0744746584445238,
|
|
"num_tokens": 495570.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.125858116149903,
|
|
"epoch": 0.21748779405237462,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.7248,
|
|
"mean_token_accuracy": 0.07628845125436783,
|
|
"num_tokens": 506103.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.083836793899536,
|
|
"epoch": 0.22192632046160674,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.6989,
|
|
"mean_token_accuracy": 0.07588708400726318,
|
|
"num_tokens": 516416.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.041877603530883,
|
|
"epoch": 0.22636484687083888,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.7498,
|
|
"mean_token_accuracy": 0.0751910775899887,
|
|
"num_tokens": 525841.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.231132793426514,
|
|
"epoch": 0.23080337328007103,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.7681,
|
|
"mean_token_accuracy": 0.07214248739182949,
|
|
"num_tokens": 536426.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.028632116317748,
|
|
"epoch": 0.23524189968930315,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7229,
|
|
"mean_token_accuracy": 0.07526009976863861,
|
|
"num_tokens": 545852.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.998026943206787,
|
|
"epoch": 0.2396804260985353,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.6466,
|
|
"mean_token_accuracy": 0.07897478640079499,
|
|
"num_tokens": 555222.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.9018083095550535,
|
|
"epoch": 0.2441189525077674,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6683,
|
|
"mean_token_accuracy": 0.0769478440284729,
|
|
"num_tokens": 566311.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.067350673675538,
|
|
"epoch": 0.24855747891699956,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.6718,
|
|
"mean_token_accuracy": 0.07673554085195064,
|
|
"num_tokens": 576883.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.010624122619628,
|
|
"epoch": 0.2529960053262317,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.8136,
|
|
"mean_token_accuracy": 0.07454071268439293,
|
|
"num_tokens": 587387.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.04631519317627,
|
|
"epoch": 0.2574345317354638,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.6914,
|
|
"mean_token_accuracy": 0.07921014353632927,
|
|
"num_tokens": 597229.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.029859972000121,
|
|
"epoch": 0.26187305814469597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.6793,
|
|
"mean_token_accuracy": 0.0733891949057579,
|
|
"num_tokens": 606997.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.9327795028686525,
|
|
"epoch": 0.2663115845539281,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.6346,
|
|
"mean_token_accuracy": 0.07976229339838029,
|
|
"num_tokens": 617316.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.945201730728149,
|
|
"epoch": 0.2707501109631602,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.6262,
|
|
"mean_token_accuracy": 0.07671116665005684,
|
|
"num_tokens": 626494.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.911632776260376,
|
|
"epoch": 0.27518863737239235,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.6211,
|
|
"mean_token_accuracy": 0.07656443342566491,
|
|
"num_tokens": 636816.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.9385381698608395,
|
|
"epoch": 0.2796271637816245,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.5921,
|
|
"mean_token_accuracy": 0.07875077873468399,
|
|
"num_tokens": 647520.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.8889247417449955,
|
|
"epoch": 0.28406569019085665,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.5615,
|
|
"mean_token_accuracy": 0.0802665926516056,
|
|
"num_tokens": 656772.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.87529935836792,
|
|
"epoch": 0.2885042166000888,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.6269,
|
|
"mean_token_accuracy": 0.07935804799199105,
|
|
"num_tokens": 666570.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.889472913742066,
|
|
"epoch": 0.2929427430093209,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.6449,
|
|
"mean_token_accuracy": 0.07449713721871376,
|
|
"num_tokens": 676873.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.901484632492066,
|
|
"epoch": 0.29738126941855303,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.556,
|
|
"mean_token_accuracy": 0.0819828025996685,
|
|
"num_tokens": 687322.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.938011121749878,
|
|
"epoch": 0.3018197958277852,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.6304,
|
|
"mean_token_accuracy": 0.07710144743323326,
|
|
"num_tokens": 697657.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.953681612014771,
|
|
"epoch": 0.3062583222370173,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.5355,
|
|
"mean_token_accuracy": 0.07943713515996934,
|
|
"num_tokens": 707713.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.818657636642456,
|
|
"epoch": 0.31069684864624947,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.5792,
|
|
"mean_token_accuracy": 0.08700495660305023,
|
|
"num_tokens": 719438.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.922421979904175,
|
|
"epoch": 0.31513537505548156,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.5211,
|
|
"mean_token_accuracy": 0.0777974933385849,
|
|
"num_tokens": 728991.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.7726140975952145,
|
|
"epoch": 0.3195739014647137,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.4375,
|
|
"mean_token_accuracy": 0.09254143610596657,
|
|
"num_tokens": 738276.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.8144529342651365,
|
|
"epoch": 0.32401242787394585,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.537,
|
|
"mean_token_accuracy": 0.07724328823387623,
|
|
"num_tokens": 748405.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.8304845809936525,
|
|
"epoch": 0.328450954283178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.5365,
|
|
"mean_token_accuracy": 0.08004771247506141,
|
|
"num_tokens": 759111.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.755885601043701,
|
|
"epoch": 0.33288948069241014,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.4431,
|
|
"mean_token_accuracy": 0.08885042890906333,
|
|
"num_tokens": 770315.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.815682411193848,
|
|
"epoch": 0.33732800710164224,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.5761,
|
|
"mean_token_accuracy": 0.07824730798602104,
|
|
"num_tokens": 781390.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.868446302413941,
|
|
"epoch": 0.3417665335108744,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.4659,
|
|
"mean_token_accuracy": 0.08218091987073421,
|
|
"num_tokens": 790710.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.664791536331177,
|
|
"epoch": 0.34620505992010653,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.4344,
|
|
"mean_token_accuracy": 0.08516876846551895,
|
|
"num_tokens": 800475.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.776907110214234,
|
|
"epoch": 0.3506435863293387,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5457,
|
|
"mean_token_accuracy": 0.07697484940290451,
|
|
"num_tokens": 810394.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.851606655120849,
|
|
"epoch": 0.3550821127385708,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.5259,
|
|
"mean_token_accuracy": 0.07952482551336289,
|
|
"num_tokens": 820493.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.881973552703857,
|
|
"epoch": 0.3595206391478029,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.5896,
|
|
"mean_token_accuracy": 0.07090779766440392,
|
|
"num_tokens": 831570.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.75118236541748,
|
|
"epoch": 0.36395916555703506,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5822,
|
|
"mean_token_accuracy": 0.07436199747025966,
|
|
"num_tokens": 841851.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.8130326747894285,
|
|
"epoch": 0.3683976919662672,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.4771,
|
|
"mean_token_accuracy": 0.08917279317975044,
|
|
"num_tokens": 851221.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.784592771530152,
|
|
"epoch": 0.37283621837549935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5136,
|
|
"mean_token_accuracy": 0.08482674360275269,
|
|
"num_tokens": 861876.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.806199550628662,
|
|
"epoch": 0.37727474478473144,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.5012,
|
|
"mean_token_accuracy": 0.08520847856998444,
|
|
"num_tokens": 871742.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.63223876953125,
|
|
"epoch": 0.3817132711939636,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.4356,
|
|
"mean_token_accuracy": 0.08614917695522309,
|
|
"num_tokens": 881968.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.708009338378906,
|
|
"epoch": 0.38615179760319573,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.3441,
|
|
"mean_token_accuracy": 0.0903160773217678,
|
|
"num_tokens": 892142.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.681766939163208,
|
|
"epoch": 0.3905903240124279,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.408,
|
|
"mean_token_accuracy": 0.09188471287488938,
|
|
"num_tokens": 901992.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.797896862030029,
|
|
"epoch": 0.39502885042166,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.5382,
|
|
"mean_token_accuracy": 0.07850045077502728,
|
|
"num_tokens": 913438.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.55986213684082,
|
|
"epoch": 0.3994673768308921,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.273,
|
|
"mean_token_accuracy": 0.10574427619576454,
|
|
"num_tokens": 923879.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.674528026580811,
|
|
"epoch": 0.40390590324012426,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.4341,
|
|
"mean_token_accuracy": 0.08904042169451713,
|
|
"num_tokens": 933954.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.707700920104981,
|
|
"epoch": 0.4083444296493564,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.4156,
|
|
"mean_token_accuracy": 0.0867860458791256,
|
|
"num_tokens": 942992.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.656433868408203,
|
|
"epoch": 0.41278295605858856,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.4528,
|
|
"mean_token_accuracy": 0.08972205594182014,
|
|
"num_tokens": 953822.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.627706384658813,
|
|
"epoch": 0.4172214824678207,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.3472,
|
|
"mean_token_accuracy": 0.08931803703308105,
|
|
"num_tokens": 964475.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.642127990722656,
|
|
"epoch": 0.4216600088770528,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.434,
|
|
"mean_token_accuracy": 0.0859565369784832,
|
|
"num_tokens": 974535.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.666657447814941,
|
|
"epoch": 0.42609853528628494,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.3944,
|
|
"mean_token_accuracy": 0.08582191392779351,
|
|
"num_tokens": 984674.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.544086360931397,
|
|
"epoch": 0.4305370616955171,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.37,
|
|
"mean_token_accuracy": 0.09044271633028984,
|
|
"num_tokens": 995248.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.742295646667481,
|
|
"epoch": 0.43497558810474923,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.3632,
|
|
"mean_token_accuracy": 0.0895160473883152,
|
|
"num_tokens": 1005547.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.550811624526977,
|
|
"epoch": 0.4394141145139814,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3413,
|
|
"mean_token_accuracy": 0.08750802129507065,
|
|
"num_tokens": 1015498.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.609345579147339,
|
|
"epoch": 0.44385264092321347,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.4055,
|
|
"mean_token_accuracy": 0.08584300205111503,
|
|
"num_tokens": 1025420.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.44385264092321347,
|
|
"eval_entropy": 7.669124412989673,
|
|
"eval_loss": 7.446435451507568,
|
|
"eval_mean_token_accuracy": 0.08312557260738132,
|
|
"eval_num_tokens": 1025420.0,
|
|
"eval_runtime": 2.7854,
|
|
"eval_samples_per_second": 1208.804,
|
|
"eval_steps_per_second": 151.145,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.6334656238555905,
|
|
"epoch": 0.4482911673324456,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.3295,
|
|
"mean_token_accuracy": 0.09022360593080521,
|
|
"num_tokens": 1036556.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.584210634231567,
|
|
"epoch": 0.45272969374167776,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.3593,
|
|
"mean_token_accuracy": 0.08703627660870553,
|
|
"num_tokens": 1046676.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.6705080509185795,
|
|
"epoch": 0.4571682201509099,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.4042,
|
|
"mean_token_accuracy": 0.08590352982282638,
|
|
"num_tokens": 1057088.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.5486232280731205,
|
|
"epoch": 0.46160674656014206,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.2982,
|
|
"mean_token_accuracy": 0.09567688480019569,
|
|
"num_tokens": 1067814.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.543769884109497,
|
|
"epoch": 0.46604527296937415,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.3327,
|
|
"mean_token_accuracy": 0.0892727330327034,
|
|
"num_tokens": 1078600.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.641923093795777,
|
|
"epoch": 0.4704837993786063,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.3363,
|
|
"mean_token_accuracy": 0.08722671940922737,
|
|
"num_tokens": 1089158.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.568871212005615,
|
|
"epoch": 0.47492232578783844,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.3497,
|
|
"mean_token_accuracy": 0.08428739383816719,
|
|
"num_tokens": 1100365.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.500961780548096,
|
|
"epoch": 0.4793608521970706,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.305,
|
|
"mean_token_accuracy": 0.09392210468649864,
|
|
"num_tokens": 1111168.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.573345041275024,
|
|
"epoch": 0.48379937860630273,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.3194,
|
|
"mean_token_accuracy": 0.09149189814925193,
|
|
"num_tokens": 1121520.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.536744737625122,
|
|
"epoch": 0.4882379050155348,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.2073,
|
|
"mean_token_accuracy": 0.09879742562770844,
|
|
"num_tokens": 1131300.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.364904880523682,
|
|
"epoch": 0.49267643142476697,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.2532,
|
|
"mean_token_accuracy": 0.0986358568072319,
|
|
"num_tokens": 1141897.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.541038084030151,
|
|
"epoch": 0.4971149578339991,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.2956,
|
|
"mean_token_accuracy": 0.08940311297774314,
|
|
"num_tokens": 1152510.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.552376794815063,
|
|
"epoch": 0.5015534842432312,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.3001,
|
|
"mean_token_accuracy": 0.09093584269285201,
|
|
"num_tokens": 1161762.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.479511404037476,
|
|
"epoch": 0.5059920106524634,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.2394,
|
|
"mean_token_accuracy": 0.09156333580613137,
|
|
"num_tokens": 1172836.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.480563068389893,
|
|
"epoch": 0.5104305370616955,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.2935,
|
|
"mean_token_accuracy": 0.09051846116781234,
|
|
"num_tokens": 1182101.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.511827182769776,
|
|
"epoch": 0.5148690634709276,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.2868,
|
|
"mean_token_accuracy": 0.08703599572181701,
|
|
"num_tokens": 1191925.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.430129528045654,
|
|
"epoch": 0.5193075898801598,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.1629,
|
|
"mean_token_accuracy": 0.09691138043999672,
|
|
"num_tokens": 1202030.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.408132934570313,
|
|
"epoch": 0.5237461162893919,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.253,
|
|
"mean_token_accuracy": 0.09141650795936584,
|
|
"num_tokens": 1212178.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.515611791610718,
|
|
"epoch": 0.5281846426986241,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.1971,
|
|
"mean_token_accuracy": 0.0950719341635704,
|
|
"num_tokens": 1221685.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.383773374557495,
|
|
"epoch": 0.5326231691078562,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.197,
|
|
"mean_token_accuracy": 0.08822897970676422,
|
|
"num_tokens": 1233178.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.4588648796081545,
|
|
"epoch": 0.5370616955170884,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.1611,
|
|
"mean_token_accuracy": 0.09191120117902755,
|
|
"num_tokens": 1243687.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.358021211624146,
|
|
"epoch": 0.5415002219263204,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.2349,
|
|
"mean_token_accuracy": 0.08815527036786079,
|
|
"num_tokens": 1255124.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.400767421722412,
|
|
"epoch": 0.5459387483355526,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.2452,
|
|
"mean_token_accuracy": 0.09174467995762825,
|
|
"num_tokens": 1265517.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.310306978225708,
|
|
"epoch": 0.5503772747447847,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.1176,
|
|
"mean_token_accuracy": 0.09506009593605995,
|
|
"num_tokens": 1275877.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.400981378555298,
|
|
"epoch": 0.5548158011540169,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.2286,
|
|
"mean_token_accuracy": 0.09438399225473404,
|
|
"num_tokens": 1285951.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.509041738510132,
|
|
"epoch": 0.559254327563249,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.1567,
|
|
"mean_token_accuracy": 0.09806277230381966,
|
|
"num_tokens": 1295209.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.316894149780273,
|
|
"epoch": 0.5636928539724811,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.2208,
|
|
"mean_token_accuracy": 0.09708264470100403,
|
|
"num_tokens": 1304733.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.342378044128418,
|
|
"epoch": 0.5681313803817133,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.1539,
|
|
"mean_token_accuracy": 0.08910495042800903,
|
|
"num_tokens": 1314981.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.538545989990235,
|
|
"epoch": 0.5725699067909454,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.198,
|
|
"mean_token_accuracy": 0.0895936667919159,
|
|
"num_tokens": 1325211.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.2863500118255615,
|
|
"epoch": 0.5770084332001776,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.1309,
|
|
"mean_token_accuracy": 0.09338074773550034,
|
|
"num_tokens": 1335625.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.436445474624634,
|
|
"epoch": 0.5814469596094097,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.2488,
|
|
"mean_token_accuracy": 0.0991661287844181,
|
|
"num_tokens": 1345158.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.262508535385132,
|
|
"epoch": 0.5858854860186418,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.081,
|
|
"mean_token_accuracy": 0.10250589922070504,
|
|
"num_tokens": 1355092.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.307907056808472,
|
|
"epoch": 0.5903240124278739,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.1329,
|
|
"mean_token_accuracy": 0.09874422326683999,
|
|
"num_tokens": 1364919.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.2338110446929935,
|
|
"epoch": 0.5947625388371061,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.1726,
|
|
"mean_token_accuracy": 0.09538876637816429,
|
|
"num_tokens": 1375200.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.331265687942505,
|
|
"epoch": 0.5992010652463382,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.1466,
|
|
"mean_token_accuracy": 0.09369650185108185,
|
|
"num_tokens": 1385728.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.4757524013519285,
|
|
"epoch": 0.6036395916555704,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.196,
|
|
"mean_token_accuracy": 0.08721103966236114,
|
|
"num_tokens": 1395542.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.264866781234741,
|
|
"epoch": 0.6080781180648025,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.1667,
|
|
"mean_token_accuracy": 0.09908052235841751,
|
|
"num_tokens": 1406350.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.250851583480835,
|
|
"epoch": 0.6125166444740346,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.0933,
|
|
"mean_token_accuracy": 0.09707043692469597,
|
|
"num_tokens": 1418229.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.257457447052002,
|
|
"epoch": 0.6169551708832668,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.101,
|
|
"mean_token_accuracy": 0.10156845226883889,
|
|
"num_tokens": 1428290.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.389108514785766,
|
|
"epoch": 0.6213936972924989,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.1644,
|
|
"mean_token_accuracy": 0.08962787240743637,
|
|
"num_tokens": 1439837.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.230832195281982,
|
|
"epoch": 0.625832223701731,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.1004,
|
|
"mean_token_accuracy": 0.09468400254845619,
|
|
"num_tokens": 1450831.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.241072797775269,
|
|
"epoch": 0.6302707501109631,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.0787,
|
|
"mean_token_accuracy": 0.0992747537791729,
|
|
"num_tokens": 1461772.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.251083326339722,
|
|
"epoch": 0.6347092765201953,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.0678,
|
|
"mean_token_accuracy": 0.10479466319084167,
|
|
"num_tokens": 1471669.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.284691286087036,
|
|
"epoch": 0.6391478029294274,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.0922,
|
|
"mean_token_accuracy": 0.0977623738348484,
|
|
"num_tokens": 1482319.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.207487869262695,
|
|
"epoch": 0.6435863293386596,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.0654,
|
|
"mean_token_accuracy": 0.09616786390542983,
|
|
"num_tokens": 1492393.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.214288234710693,
|
|
"epoch": 0.6480248557478917,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.1377,
|
|
"mean_token_accuracy": 0.10020551234483718,
|
|
"num_tokens": 1502986.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.309820938110351,
|
|
"epoch": 0.6524633821571239,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.1195,
|
|
"mean_token_accuracy": 0.09985571056604385,
|
|
"num_tokens": 1514046.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.249904489517212,
|
|
"epoch": 0.656901908566356,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.133,
|
|
"mean_token_accuracy": 0.0978231132030487,
|
|
"num_tokens": 1523789.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.303426790237427,
|
|
"epoch": 0.6613404349755881,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.0636,
|
|
"mean_token_accuracy": 0.0984602726995945,
|
|
"num_tokens": 1533449.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.157350540161133,
|
|
"epoch": 0.6657789613848203,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.1052,
|
|
"mean_token_accuracy": 0.0913457877933979,
|
|
"num_tokens": 1543449.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.232370662689209,
|
|
"epoch": 0.6702174877940523,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.957,
|
|
"mean_token_accuracy": 0.10332913659512996,
|
|
"num_tokens": 1553529.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.166843461990356,
|
|
"epoch": 0.6746560142032845,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.0559,
|
|
"mean_token_accuracy": 0.1009259507060051,
|
|
"num_tokens": 1564245.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.135752153396607,
|
|
"epoch": 0.6790945406125166,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.9787,
|
|
"mean_token_accuracy": 0.0997254841029644,
|
|
"num_tokens": 1573834.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.034980821609497,
|
|
"epoch": 0.6835330670217488,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.9407,
|
|
"mean_token_accuracy": 0.10062595605850219,
|
|
"num_tokens": 1583782.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.211769485473633,
|
|
"epoch": 0.6879715934309809,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.0625,
|
|
"mean_token_accuracy": 0.10229107290506363,
|
|
"num_tokens": 1594056.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.17944564819336,
|
|
"epoch": 0.6924101198402131,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.0671,
|
|
"mean_token_accuracy": 0.09898089095950127,
|
|
"num_tokens": 1604177.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.1694622993469235,
|
|
"epoch": 0.6968486462494452,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.9696,
|
|
"mean_token_accuracy": 0.10202183946967125,
|
|
"num_tokens": 1614439.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.079647397994995,
|
|
"epoch": 0.7012871726586773,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.99,
|
|
"mean_token_accuracy": 0.09851381108164788,
|
|
"num_tokens": 1625004.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.182702732086182,
|
|
"epoch": 0.7057256990679095,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.0091,
|
|
"mean_token_accuracy": 0.09595384523272514,
|
|
"num_tokens": 1634188.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.0714222431182865,
|
|
"epoch": 0.7101642254771416,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.9157,
|
|
"mean_token_accuracy": 0.1082749105989933,
|
|
"num_tokens": 1642633.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.0621692657470705,
|
|
"epoch": 0.7146027518863737,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.9424,
|
|
"mean_token_accuracy": 0.10712620615959167,
|
|
"num_tokens": 1654536.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.148227691650391,
|
|
"epoch": 0.7190412782956058,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.0822,
|
|
"mean_token_accuracy": 0.09017667174339294,
|
|
"num_tokens": 1665083.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.069796466827393,
|
|
"epoch": 0.723479804704838,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.033,
|
|
"mean_token_accuracy": 0.09541632980108261,
|
|
"num_tokens": 1675826.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.110815143585205,
|
|
"epoch": 0.7279183311140701,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.9946,
|
|
"mean_token_accuracy": 0.10398527979850769,
|
|
"num_tokens": 1685806.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.097141170501709,
|
|
"epoch": 0.7323568575233023,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.0632,
|
|
"mean_token_accuracy": 0.10135304108262062,
|
|
"num_tokens": 1695984.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.124357271194458,
|
|
"epoch": 0.7367953839325344,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.0575,
|
|
"mean_token_accuracy": 0.09942566752433776,
|
|
"num_tokens": 1706086.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.091458511352539,
|
|
"epoch": 0.7412339103417666,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.0284,
|
|
"mean_token_accuracy": 0.10059899911284446,
|
|
"num_tokens": 1718040.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.131758165359497,
|
|
"epoch": 0.7456724367509987,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.0191,
|
|
"mean_token_accuracy": 0.09906580075621604,
|
|
"num_tokens": 1729009.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.087246465682983,
|
|
"epoch": 0.7501109631602308,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.9929,
|
|
"mean_token_accuracy": 0.10902519077062607,
|
|
"num_tokens": 1738895.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.113551664352417,
|
|
"epoch": 0.7545494895694629,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.0261,
|
|
"mean_token_accuracy": 0.10415812507271767,
|
|
"num_tokens": 1750225.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.0047619342803955,
|
|
"epoch": 0.758988015978695,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.9292,
|
|
"mean_token_accuracy": 0.105811557918787,
|
|
"num_tokens": 1761224.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.085004186630249,
|
|
"epoch": 0.7634265423879272,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.09775067865848541,
|
|
"num_tokens": 1771853.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.146742153167724,
|
|
"epoch": 0.7678650687971593,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.9895,
|
|
"mean_token_accuracy": 0.10439399853348733,
|
|
"num_tokens": 1781186.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.151022291183471,
|
|
"epoch": 0.7723035952063915,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.9711,
|
|
"mean_token_accuracy": 0.09423294179141521,
|
|
"num_tokens": 1792887.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.941896152496338,
|
|
"epoch": 0.7767421216156236,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.8708,
|
|
"mean_token_accuracy": 0.10734946057200431,
|
|
"num_tokens": 1803050.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.811341190338135,
|
|
"epoch": 0.7811806480248558,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.9879,
|
|
"mean_token_accuracy": 0.10939516723155976,
|
|
"num_tokens": 1813287.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.056670713424682,
|
|
"epoch": 0.7856191744340879,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9735,
|
|
"mean_token_accuracy": 0.1008478730916977,
|
|
"num_tokens": 1824425.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.97848916053772,
|
|
"epoch": 0.79005770084332,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8898,
|
|
"mean_token_accuracy": 0.11324303895235062,
|
|
"num_tokens": 1833968.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.980572080612182,
|
|
"epoch": 0.7944962272525522,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.0128,
|
|
"mean_token_accuracy": 0.10274540036916732,
|
|
"num_tokens": 1844097.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.055815601348877,
|
|
"epoch": 0.7989347536617842,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.934,
|
|
"mean_token_accuracy": 0.10776300206780434,
|
|
"num_tokens": 1854395.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.931429195404053,
|
|
"epoch": 0.8033732800710164,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.949,
|
|
"mean_token_accuracy": 0.10048572197556496,
|
|
"num_tokens": 1864531.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.998499774932862,
|
|
"epoch": 0.8078118064802485,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8224,
|
|
"mean_token_accuracy": 0.11029603257775307,
|
|
"num_tokens": 1875071.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.971027994155884,
|
|
"epoch": 0.8122503328894807,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.947,
|
|
"mean_token_accuracy": 0.10487593412399292,
|
|
"num_tokens": 1884633.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.975761222839355,
|
|
"epoch": 0.8166888592987128,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.9402,
|
|
"mean_token_accuracy": 0.10628680288791656,
|
|
"num_tokens": 1895204.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.054645633697509,
|
|
"epoch": 0.821127385707945,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.9658,
|
|
"mean_token_accuracy": 0.10290167108178139,
|
|
"num_tokens": 1906187.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.008241748809814,
|
|
"epoch": 0.8255659121171771,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.9215,
|
|
"mean_token_accuracy": 0.09948821216821671,
|
|
"num_tokens": 1915585.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.954188346862793,
|
|
"epoch": 0.8300044385264093,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.9624,
|
|
"mean_token_accuracy": 0.1034872155636549,
|
|
"num_tokens": 1926912.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.975879383087158,
|
|
"epoch": 0.8344429649356414,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.8828,
|
|
"mean_token_accuracy": 0.1049314372241497,
|
|
"num_tokens": 1936542.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.032291650772095,
|
|
"epoch": 0.8388814913448736,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.8869,
|
|
"mean_token_accuracy": 0.1079288512468338,
|
|
"num_tokens": 1946052.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.855350923538208,
|
|
"epoch": 0.8433200177541056,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.9479,
|
|
"mean_token_accuracy": 0.10810579061508178,
|
|
"num_tokens": 1957352.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.012834692001343,
|
|
"epoch": 0.8477585441633377,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.9165,
|
|
"mean_token_accuracy": 0.10588640198111535,
|
|
"num_tokens": 1967538.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.953670644760132,
|
|
"epoch": 0.8521970705725699,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.9437,
|
|
"mean_token_accuracy": 0.10475035384297371,
|
|
"num_tokens": 1977620.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.001286792755127,
|
|
"epoch": 0.856635596981802,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8474,
|
|
"mean_token_accuracy": 0.10972543135285377,
|
|
"num_tokens": 1987534.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.883567142486572,
|
|
"epoch": 0.8610741233910342,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.8677,
|
|
"mean_token_accuracy": 0.11282802000641823,
|
|
"num_tokens": 1997817.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.960841703414917,
|
|
"epoch": 0.8655126498002663,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.8693,
|
|
"mean_token_accuracy": 0.10804260671138763,
|
|
"num_tokens": 2007529.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.893871688842774,
|
|
"epoch": 0.8699511762094985,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.8747,
|
|
"mean_token_accuracy": 0.1100342482328415,
|
|
"num_tokens": 2017317.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.920565032958985,
|
|
"epoch": 0.8743897026187306,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.8925,
|
|
"mean_token_accuracy": 0.10560473203659057,
|
|
"num_tokens": 2028427.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.928885173797608,
|
|
"epoch": 0.8788282290279628,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.8469,
|
|
"mean_token_accuracy": 0.111257666349411,
|
|
"num_tokens": 2038464.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.895605516433716,
|
|
"epoch": 0.8832667554371949,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.9248,
|
|
"mean_token_accuracy": 0.10252941846847534,
|
|
"num_tokens": 2049006.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.840125465393067,
|
|
"epoch": 0.8877052818464269,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.8673,
|
|
"mean_token_accuracy": 0.10938069224357605,
|
|
"num_tokens": 2060436.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8877052818464269,
|
|
"eval_entropy": 6.823905027394057,
|
|
"eval_loss": 6.903217315673828,
|
|
"eval_mean_token_accuracy": 0.10246794701790583,
|
|
"eval_num_tokens": 2060436.0,
|
|
"eval_runtime": 2.7045,
|
|
"eval_samples_per_second": 1244.948,
|
|
"eval_steps_per_second": 155.665,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.8510839462280275,
|
|
"epoch": 0.8921438082556591,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999998312369076,
|
|
"loss": 6.7705,
|
|
"mean_token_accuracy": 0.1085792139172554,
|
|
"num_tokens": 2070906.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.848130416870117,
|
|
"epoch": 0.8965823346648912,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999991456372788,
|
|
"loss": 6.8232,
|
|
"mean_token_accuracy": 0.10221576392650604,
|
|
"num_tokens": 2080563.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.826877164840698,
|
|
"epoch": 0.9010208610741234,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499997932655026,
|
|
"loss": 6.7509,
|
|
"mean_token_accuracy": 0.11835153996944428,
|
|
"num_tokens": 2089785.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.89136643409729,
|
|
"epoch": 0.9054593874833555,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999961922929925,
|
|
"loss": 6.798,
|
|
"mean_token_accuracy": 0.11268565952777862,
|
|
"num_tokens": 2099609.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.894064235687256,
|
|
"epoch": 0.9098979138925877,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999939245552573,
|
|
"loss": 6.8422,
|
|
"mean_token_accuracy": 0.11023318842053413,
|
|
"num_tokens": 2110043.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.8520002365112305,
|
|
"epoch": 0.9143364403018198,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499991129447136,
|
|
"loss": 6.8665,
|
|
"mean_token_accuracy": 0.1076730340719223,
|
|
"num_tokens": 2119748.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.8474874019622805,
|
|
"epoch": 0.918774966711052,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999878069751803,
|
|
"loss": 6.8907,
|
|
"mean_token_accuracy": 0.10536454170942307,
|
|
"num_tokens": 2130347.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.900251626968384,
|
|
"epoch": 0.9232134931202841,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999839571471776,
|
|
"loss": 6.7778,
|
|
"mean_token_accuracy": 0.1090541049838066,
|
|
"num_tokens": 2140499.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.788918542861938,
|
|
"epoch": 0.9276520195295161,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999795799721517,
|
|
"loss": 6.7795,
|
|
"mean_token_accuracy": 0.10508258789777755,
|
|
"num_tokens": 2151143.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.702548313140869,
|
|
"epoch": 0.9320905459387483,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999746754603624,
|
|
"loss": 6.8099,
|
|
"mean_token_accuracy": 0.1090174950659275,
|
|
"num_tokens": 2161651.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.934295988082885,
|
|
"epoch": 0.9365290723479804,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999692436233055,
|
|
"loss": 6.8146,
|
|
"mean_token_accuracy": 0.1094007097184658,
|
|
"num_tokens": 2172152.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.8816564083099365,
|
|
"epoch": 0.9409675987572126,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499963284473713,
|
|
"loss": 6.8335,
|
|
"mean_token_accuracy": 0.1026708424091339,
|
|
"num_tokens": 2182912.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.725460863113403,
|
|
"epoch": 0.9454061251664447,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999567980255526,
|
|
"loss": 6.8253,
|
|
"mean_token_accuracy": 0.10950745418667793,
|
|
"num_tokens": 2193573.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.998191976547242,
|
|
"epoch": 0.9498446515756769,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999497842940282,
|
|
"loss": 6.8627,
|
|
"mean_token_accuracy": 0.1091019332408905,
|
|
"num_tokens": 2204157.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.804390239715576,
|
|
"epoch": 0.954283177984909,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999422432955794,
|
|
"loss": 6.8426,
|
|
"mean_token_accuracy": 0.11238477602601052,
|
|
"num_tokens": 2214433.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.7941382884979244,
|
|
"epoch": 0.9587217043941412,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999341750478818,
|
|
"loss": 6.7722,
|
|
"mean_token_accuracy": 0.1097018837928772,
|
|
"num_tokens": 2224913.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.810701084136963,
|
|
"epoch": 0.9631602308033733,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499925579569847,
|
|
"loss": 6.7335,
|
|
"mean_token_accuracy": 0.12259945124387742,
|
|
"num_tokens": 2234871.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.818542575836181,
|
|
"epoch": 0.9675987572126055,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004999164568816219,
|
|
"loss": 6.7889,
|
|
"mean_token_accuracy": 0.10602138787508011,
|
|
"num_tokens": 2244462.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.706165170669555,
|
|
"epoch": 0.9720372836218375,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999068070045897,
|
|
"loss": 6.7827,
|
|
"mean_token_accuracy": 0.10748272910714149,
|
|
"num_tokens": 2256145.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.83993878364563,
|
|
"epoch": 0.9764758100310696,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499896629961369,
|
|
"loss": 6.7837,
|
|
"mean_token_accuracy": 0.11425440460443496,
|
|
"num_tokens": 2265913.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.725751304626465,
|
|
"epoch": 0.9809143364403018,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499885925775814,
|
|
"loss": 6.7316,
|
|
"mean_token_accuracy": 0.11490511074662209,
|
|
"num_tokens": 2274874.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.830372858047485,
|
|
"epoch": 0.9853528628495339,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998746944730148,
|
|
"loss": 6.7131,
|
|
"mean_token_accuracy": 0.11283201426267624,
|
|
"num_tokens": 2284767.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.874453449249268,
|
|
"epoch": 0.9897913892587661,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.8831,
|
|
"mean_token_accuracy": 0.10844636484980583,
|
|
"num_tokens": 2295196.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.715284633636474,
|
|
"epoch": 0.9942299156679982,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998506506222202,
|
|
"loss": 6.7286,
|
|
"mean_token_accuracy": 0.11680992171168328,
|
|
"num_tokens": 2306078.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.814456415176392,
|
|
"epoch": 0.9986684420772304,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998378381305821,
|
|
"loss": 6.8094,
|
|
"mean_token_accuracy": 0.10992804765701295,
|
|
"num_tokens": 2317229.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.859947204589844,
|
|
"epoch": 1.0026631158455392,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998244986344138,
|
|
"loss": 6.688,
|
|
"mean_token_accuracy": 0.11234754075606664,
|
|
"num_tokens": 2326312.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.762184476852417,
|
|
"epoch": 1.0071016422547714,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998106321649822,
|
|
"loss": 6.5436,
|
|
"mean_token_accuracy": 0.12012127339839936,
|
|
"num_tokens": 2335548.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.783463954925537,
|
|
"epoch": 1.0115401686640035,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997962387547893,
|
|
"loss": 6.5731,
|
|
"mean_token_accuracy": 0.11604165062308311,
|
|
"num_tokens": 2346180.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.728586292266845,
|
|
"epoch": 1.0159786950732357,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997813184375723,
|
|
"loss": 6.5403,
|
|
"mean_token_accuracy": 0.11700786501169205,
|
|
"num_tokens": 2355682.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.737982702255249,
|
|
"epoch": 1.0204172214824678,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997658712483034,
|
|
"loss": 6.5789,
|
|
"mean_token_accuracy": 0.11569899022579193,
|
|
"num_tokens": 2365529.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.645860004425049,
|
|
"epoch": 1.0248557478917,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997498972231898,
|
|
"loss": 6.4906,
|
|
"mean_token_accuracy": 0.12428833544254303,
|
|
"num_tokens": 2375527.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.723867797851563,
|
|
"epoch": 1.0292942743009321,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997333963996734,
|
|
"loss": 6.6103,
|
|
"mean_token_accuracy": 0.11796300113201141,
|
|
"num_tokens": 2386425.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.686568021774292,
|
|
"epoch": 1.0337328007101643,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997163688164313,
|
|
"loss": 6.6209,
|
|
"mean_token_accuracy": 0.11397269815206527,
|
|
"num_tokens": 2396897.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.736155891418457,
|
|
"epoch": 1.0381713271193964,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996988145133745,
|
|
"loss": 6.6215,
|
|
"mean_token_accuracy": 0.11120860427618026,
|
|
"num_tokens": 2408671.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.68158974647522,
|
|
"epoch": 1.0426098535286286,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996807335316496,
|
|
"loss": 6.5968,
|
|
"mean_token_accuracy": 0.11955826655030251,
|
|
"num_tokens": 2418373.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.73370852470398,
|
|
"epoch": 1.0470483799378607,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996621259136368,
|
|
"loss": 6.5516,
|
|
"mean_token_accuracy": 0.12124700546264648,
|
|
"num_tokens": 2429094.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.58090991973877,
|
|
"epoch": 1.0514869063470929,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996429917029513,
|
|
"loss": 6.5521,
|
|
"mean_token_accuracy": 0.11974199712276459,
|
|
"num_tokens": 2439522.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.689072942733764,
|
|
"epoch": 1.055925432756325,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996233309444424,
|
|
"loss": 6.5334,
|
|
"mean_token_accuracy": 0.12693726643919945,
|
|
"num_tokens": 2449900.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.648254156112671,
|
|
"epoch": 1.060363959165557,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996031436841934,
|
|
"loss": 6.528,
|
|
"mean_token_accuracy": 0.11835195198655128,
|
|
"num_tokens": 2461187.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.613636541366577,
|
|
"epoch": 1.064802485574789,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004995824299695219,
|
|
"loss": 6.5522,
|
|
"mean_token_accuracy": 0.12386145591735839,
|
|
"num_tokens": 2471897.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.695000886917114,
|
|
"epoch": 1.0692410119840212,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995611898489796,
|
|
"loss": 6.4835,
|
|
"mean_token_accuracy": 0.13428002893924712,
|
|
"num_tokens": 2482668.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.542116737365722,
|
|
"epoch": 1.0736795383932534,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995394233723516,
|
|
"loss": 6.5273,
|
|
"mean_token_accuracy": 0.12380995452404023,
|
|
"num_tokens": 2492375.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.612301158905029,
|
|
"epoch": 1.0781180648024855,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004995171305906574,
|
|
"loss": 6.5553,
|
|
"mean_token_accuracy": 0.12436272650957107,
|
|
"num_tokens": 2504176.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.712629508972168,
|
|
"epoch": 1.0825565912117177,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004994943115561495,
|
|
"loss": 6.5802,
|
|
"mean_token_accuracy": 0.12135286405682563,
|
|
"num_tokens": 2513437.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.71323561668396,
|
|
"epoch": 1.0869951176209498,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004994709663223143,
|
|
"loss": 6.5305,
|
|
"mean_token_accuracy": 0.12117772102355957,
|
|
"num_tokens": 2524907.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.586472082138061,
|
|
"epoch": 1.091433644030182,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994470949438712,
|
|
"loss": 6.4818,
|
|
"mean_token_accuracy": 0.1203920915722847,
|
|
"num_tokens": 2535288.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.5804437637329105,
|
|
"epoch": 1.095872170439414,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004994226974767734,
|
|
"loss": 6.5353,
|
|
"mean_token_accuracy": 0.11931732892990113,
|
|
"num_tokens": 2545896.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.543254709243774,
|
|
"epoch": 1.1003106968486462,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993977739782066,
|
|
"loss": 6.4584,
|
|
"mean_token_accuracy": 0.13561398088932036,
|
|
"num_tokens": 2555440.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.59882869720459,
|
|
"epoch": 1.1047492232578784,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00049937232450659,
|
|
"loss": 6.4693,
|
|
"mean_token_accuracy": 0.12519273459911345,
|
|
"num_tokens": 2565474.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.621742391586304,
|
|
"epoch": 1.1091877496671105,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004993463491215753,
|
|
"loss": 6.5707,
|
|
"mean_token_accuracy": 0.11752136424183846,
|
|
"num_tokens": 2576090.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.66785740852356,
|
|
"epoch": 1.1136262760763427,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499319847884047,
|
|
"loss": 6.48,
|
|
"mean_token_accuracy": 0.12498680129647255,
|
|
"num_tokens": 2585719.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.547177171707153,
|
|
"epoch": 1.1180648024855748,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992928208561224,
|
|
"loss": 6.4818,
|
|
"mean_token_accuracy": 0.12778932899236678,
|
|
"num_tokens": 2594797.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.544874048233032,
|
|
"epoch": 1.122503328894807,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004992652681011508,
|
|
"loss": 6.5028,
|
|
"mean_token_accuracy": 0.12365371584892274,
|
|
"num_tokens": 2605525.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.667212867736817,
|
|
"epoch": 1.1269418553040391,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000499237189683714,
|
|
"loss": 6.543,
|
|
"mean_token_accuracy": 0.12363962829113007,
|
|
"num_tokens": 2616478.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.585494756698608,
|
|
"epoch": 1.1313803817132713,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992085856696259,
|
|
"loss": 6.5255,
|
|
"mean_token_accuracy": 0.12803056016564368,
|
|
"num_tokens": 2626290.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.660335063934326,
|
|
"epoch": 1.1358189081225034,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991794561259325,
|
|
"loss": 6.5861,
|
|
"mean_token_accuracy": 0.11459894254803657,
|
|
"num_tokens": 2637319.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.654719829559326,
|
|
"epoch": 1.1402574345317356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991498011209112,
|
|
"loss": 6.5036,
|
|
"mean_token_accuracy": 0.1196965254843235,
|
|
"num_tokens": 2648140.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.520665168762207,
|
|
"epoch": 1.1446959609409677,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991196207240714,
|
|
"loss": 6.5534,
|
|
"mean_token_accuracy": 0.12024708986282348,
|
|
"num_tokens": 2658561.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.740872001647949,
|
|
"epoch": 1.1491344873501999,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004990889150061541,
|
|
"loss": 6.421,
|
|
"mean_token_accuracy": 0.13026004359126092,
|
|
"num_tokens": 2670233.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.636859560012818,
|
|
"epoch": 1.1535730137594318,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004990576840391312,
|
|
"loss": 6.4896,
|
|
"mean_token_accuracy": 0.12343377694487571,
|
|
"num_tokens": 2681047.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.514160776138306,
|
|
"epoch": 1.158011540168664,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.4793,
|
|
"mean_token_accuracy": 0.12569797188043594,
|
|
"num_tokens": 2691284.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.50507082939148,
|
|
"epoch": 1.162450066577896,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004989936466518127,
|
|
"loss": 6.4791,
|
|
"mean_token_accuracy": 0.12733488231897355,
|
|
"num_tokens": 2702105.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.660071849822998,
|
|
"epoch": 1.1668885929871282,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989608403816164,
|
|
"loss": 6.4392,
|
|
"mean_token_accuracy": 0.1262727789580822,
|
|
"num_tokens": 2712003.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.480842781066895,
|
|
"epoch": 1.1713271193963604,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004989275091625126,
|
|
"loss": 6.4644,
|
|
"mean_token_accuracy": 0.12619971707463265,
|
|
"num_tokens": 2721609.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.514829921722412,
|
|
"epoch": 1.1757656458055925,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004988936530726276,
|
|
"loss": 6.3574,
|
|
"mean_token_accuracy": 0.1336055465042591,
|
|
"num_tokens": 2731036.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.577421188354492,
|
|
"epoch": 1.1802041722148247,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004988592721913176,
|
|
"loss": 6.4856,
|
|
"mean_token_accuracy": 0.12642226815223695,
|
|
"num_tokens": 2741860.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.4255718231201175,
|
|
"epoch": 1.1846426986240568,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004988243665991692,
|
|
"loss": 6.4757,
|
|
"mean_token_accuracy": 0.12892607748508453,
|
|
"num_tokens": 2751898.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.610957288742066,
|
|
"epoch": 1.189081225033289,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987889363779985,
|
|
"loss": 6.4425,
|
|
"mean_token_accuracy": 0.13178927153348924,
|
|
"num_tokens": 2761193.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.46262493133545,
|
|
"epoch": 1.193519751442521,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004987529816108517,
|
|
"loss": 6.4799,
|
|
"mean_token_accuracy": 0.1273398607969284,
|
|
"num_tokens": 2771693.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.6233940601348875,
|
|
"epoch": 1.1979582778517532,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004987165023820043,
|
|
"loss": 6.4757,
|
|
"mean_token_accuracy": 0.12488154098391532,
|
|
"num_tokens": 2780680.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.422232627868652,
|
|
"epoch": 1.2023968042609854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004986794987769611,
|
|
"loss": 6.4691,
|
|
"mean_token_accuracy": 0.13016403540968896,
|
|
"num_tokens": 2792049.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.61912989616394,
|
|
"epoch": 1.2068353306702175,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000498641970882456,
|
|
"loss": 6.5094,
|
|
"mean_token_accuracy": 0.11824153885245323,
|
|
"num_tokens": 2802284.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.5044067859649655,
|
|
"epoch": 1.2112738570794497,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986039187864518,
|
|
"loss": 6.4687,
|
|
"mean_token_accuracy": 0.1283252492547035,
|
|
"num_tokens": 2812791.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.563082885742188,
|
|
"epoch": 1.2157123834886818,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004985653425781401,
|
|
"loss": 6.5185,
|
|
"mean_token_accuracy": 0.12203285768628121,
|
|
"num_tokens": 2823216.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.505474853515625,
|
|
"epoch": 1.220150909897914,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004985262423479408,
|
|
"loss": 6.4274,
|
|
"mean_token_accuracy": 0.13895628824830056,
|
|
"num_tokens": 2832582.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.52580771446228,
|
|
"epoch": 1.2245894363071461,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004984866181875021,
|
|
"loss": 6.4579,
|
|
"mean_token_accuracy": 0.12877390310168266,
|
|
"num_tokens": 2843576.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.516999197006226,
|
|
"epoch": 1.229027962716378,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004984464701897005,
|
|
"loss": 6.456,
|
|
"mean_token_accuracy": 0.12886594235897064,
|
|
"num_tokens": 2854028.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.550729846954345,
|
|
"epoch": 1.2334664891256102,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004984057984486402,
|
|
"loss": 6.4491,
|
|
"mean_token_accuracy": 0.1290650874376297,
|
|
"num_tokens": 2863695.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.497787046432495,
|
|
"epoch": 1.2379050155348423,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983646030596527,
|
|
"loss": 6.4078,
|
|
"mean_token_accuracy": 0.12960500344634057,
|
|
"num_tokens": 2873134.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.530357122421265,
|
|
"epoch": 1.2423435419440745,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004983228841192975,
|
|
"loss": 6.4504,
|
|
"mean_token_accuracy": 0.12851642668247223,
|
|
"num_tokens": 2883559.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.488299894332886,
|
|
"epoch": 1.2467820683533066,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004982806417253607,
|
|
"loss": 6.422,
|
|
"mean_token_accuracy": 0.12727490290999413,
|
|
"num_tokens": 2893541.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.5642862796783445,
|
|
"epoch": 1.2512205947625388,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004982378759768557,
|
|
"loss": 6.459,
|
|
"mean_token_accuracy": 0.12929099127650262,
|
|
"num_tokens": 2903390.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.462419414520264,
|
|
"epoch": 1.255659121171771,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004981945869740225,
|
|
"loss": 6.3934,
|
|
"mean_token_accuracy": 0.13202970251441,
|
|
"num_tokens": 2913071.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.547803783416748,
|
|
"epoch": 1.260097647581003,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004981507748183276,
|
|
"loss": 6.5538,
|
|
"mean_token_accuracy": 0.12963883131742476,
|
|
"num_tokens": 2925266.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.489030170440674,
|
|
"epoch": 1.2645361739902352,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981064396124635,
|
|
"loss": 6.4396,
|
|
"mean_token_accuracy": 0.12705525755882263,
|
|
"num_tokens": 2935347.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.553302574157715,
|
|
"epoch": 1.2689747003994674,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004980615814603492,
|
|
"loss": 6.5228,
|
|
"mean_token_accuracy": 0.12255004495382309,
|
|
"num_tokens": 2945381.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.548076343536377,
|
|
"epoch": 1.2734132268086995,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004980162004671288,
|
|
"loss": 6.4853,
|
|
"mean_token_accuracy": 0.12323620095849037,
|
|
"num_tokens": 2956332.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.490498161315918,
|
|
"epoch": 1.2778517532179317,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004979702967391725,
|
|
"loss": 6.4637,
|
|
"mean_token_accuracy": 0.1297840654850006,
|
|
"num_tokens": 2966563.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.459001588821411,
|
|
"epoch": 1.2822902796271638,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004979238703840755,
|
|
"loss": 6.4291,
|
|
"mean_token_accuracy": 0.13188086450099945,
|
|
"num_tokens": 2976078.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.535675382614135,
|
|
"epoch": 1.286728806036396,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004978769215106579,
|
|
"loss": 6.51,
|
|
"mean_token_accuracy": 0.11516684889793397,
|
|
"num_tokens": 2985889.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.561489534378052,
|
|
"epoch": 1.291167332445628,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004978294502289646,
|
|
"loss": 6.4789,
|
|
"mean_token_accuracy": 0.12259078919887542,
|
|
"num_tokens": 2996244.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.490956830978393,
|
|
"epoch": 1.2956058588548602,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004977814566502651,
|
|
"loss": 6.4193,
|
|
"mean_token_accuracy": 0.12861331328749656,
|
|
"num_tokens": 3006212.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.499041128158569,
|
|
"epoch": 1.3000443852640924,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977329408870532,
|
|
"loss": 6.409,
|
|
"mean_token_accuracy": 0.12722128033638,
|
|
"num_tokens": 3016763.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.4857086658477785,
|
|
"epoch": 1.3044829116733245,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004976839030530464,
|
|
"loss": 6.3783,
|
|
"mean_token_accuracy": 0.13740343004465103,
|
|
"num_tokens": 3027111.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.542345571517944,
|
|
"epoch": 1.3089214380825567,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497634343263186,
|
|
"loss": 6.4623,
|
|
"mean_token_accuracy": 0.12753049805760383,
|
|
"num_tokens": 3037549.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.353925800323486,
|
|
"epoch": 1.3133599644917888,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004975842616336369,
|
|
"loss": 6.386,
|
|
"mean_token_accuracy": 0.13300835117697715,
|
|
"num_tokens": 3048033.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.501514005661011,
|
|
"epoch": 1.317798490901021,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004975336582817869,
|
|
"loss": 6.3997,
|
|
"mean_token_accuracy": 0.13204055801033973,
|
|
"num_tokens": 3057503.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.487248039245605,
|
|
"epoch": 1.3222370173102531,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004974825333262469,
|
|
"loss": 6.4642,
|
|
"mean_token_accuracy": 0.12323634922504426,
|
|
"num_tokens": 3068216.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.558629322052002,
|
|
"epoch": 1.3266755437194853,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 6.4271,
|
|
"mean_token_accuracy": 0.13236989378929137,
|
|
"num_tokens": 3079471.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.5367296695709225,
|
|
"epoch": 1.3311140701287172,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004973787190846524,
|
|
"loss": 6.518,
|
|
"mean_token_accuracy": 0.12429615929722786,
|
|
"num_tokens": 3090996.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3311140701287172,
|
|
"eval_entropy": 6.421620451639497,
|
|
"eval_loss": 6.5496826171875,
|
|
"eval_mean_token_accuracy": 0.12521972324955774,
|
|
"eval_num_tokens": 3090996.0,
|
|
"eval_runtime": 2.6959,
|
|
"eval_samples_per_second": 1248.929,
|
|
"eval_steps_per_second": 156.163,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.472316646575928,
|
|
"epoch": 1.3355525965379493,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004973260300419316,
|
|
"loss": 6.466,
|
|
"mean_token_accuracy": 0.1273823134601116,
|
|
"num_tokens": 3100445.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.405650186538696,
|
|
"epoch": 1.3399911229471815,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004972728198821871,
|
|
"loss": 6.3772,
|
|
"mean_token_accuracy": 0.13720163628458976,
|
|
"num_tokens": 3110065.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.48620924949646,
|
|
"epoch": 1.3444296493564136,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00049721908873014,
|
|
"loss": 6.4228,
|
|
"mean_token_accuracy": 0.12824928835034372,
|
|
"num_tokens": 3120323.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.523930549621582,
|
|
"epoch": 1.3488681757656458,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004971648367117323,
|
|
"loss": 6.464,
|
|
"mean_token_accuracy": 0.12202735692262649,
|
|
"num_tokens": 3131268.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.500436449050904,
|
|
"epoch": 1.353306702174878,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004971100639541271,
|
|
"loss": 6.3697,
|
|
"mean_token_accuracy": 0.13300676867365838,
|
|
"num_tokens": 3141529.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.385044431686401,
|
|
"epoch": 1.35774522858411,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497054770585708,
|
|
"loss": 6.3884,
|
|
"mean_token_accuracy": 0.13427990674972534,
|
|
"num_tokens": 3151096.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.540951824188232,
|
|
"epoch": 1.3621837549933422,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004969989567360788,
|
|
"loss": 6.3651,
|
|
"mean_token_accuracy": 0.126102414727211,
|
|
"num_tokens": 3160900.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.315720701217652,
|
|
"epoch": 1.3666222814025744,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004969426225360635,
|
|
"loss": 6.3444,
|
|
"mean_token_accuracy": 0.1334808275103569,
|
|
"num_tokens": 3171303.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.431712341308594,
|
|
"epoch": 1.3710608078118065,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0004968857681177056,
|
|
"loss": 6.4942,
|
|
"mean_token_accuracy": 0.12975035086274148,
|
|
"num_tokens": 3181924.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.435882043838501,
|
|
"epoch": 1.3754993342210386,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004968283936142679,
|
|
"loss": 6.359,
|
|
"mean_token_accuracy": 0.13290246427059174,
|
|
"num_tokens": 3192018.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.475548458099365,
|
|
"epoch": 1.3799378606302708,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004967704991602322,
|
|
"loss": 6.3652,
|
|
"mean_token_accuracy": 0.1294732205569744,
|
|
"num_tokens": 3203064.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.439318323135376,
|
|
"epoch": 1.384376387039503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004967120848912995,
|
|
"loss": 6.3961,
|
|
"mean_token_accuracy": 0.12799521535634995,
|
|
"num_tokens": 3213651.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.372346305847168,
|
|
"epoch": 1.388814913448735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004966531509443884,
|
|
"loss": 6.3479,
|
|
"mean_token_accuracy": 0.13649084344506263,
|
|
"num_tokens": 3224152.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.471084785461426,
|
|
"epoch": 1.3932534398579672,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004965936974576363,
|
|
"loss": 6.3545,
|
|
"mean_token_accuracy": 0.13298841714859008,
|
|
"num_tokens": 3234185.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.455093669891357,
|
|
"epoch": 1.3976919662671992,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004965337245703981,
|
|
"loss": 6.3767,
|
|
"mean_token_accuracy": 0.12946312502026558,
|
|
"num_tokens": 3244509.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.458269500732422,
|
|
"epoch": 1.4021304926764313,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004964732324232462,
|
|
"loss": 6.4008,
|
|
"mean_token_accuracy": 0.12682827115058898,
|
|
"num_tokens": 3255157.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.34567403793335,
|
|
"epoch": 1.4065690190856635,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00049641222115797,
|
|
"loss": 6.2956,
|
|
"mean_token_accuracy": 0.13407586440443992,
|
|
"num_tokens": 3264367.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.4444191455841064,
|
|
"epoch": 1.4110075454948956,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004963506909175758,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.12875387147068978,
|
|
"num_tokens": 3274678.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.426648998260498,
|
|
"epoch": 1.4154460719041277,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000496288641846286,
|
|
"loss": 6.3707,
|
|
"mean_token_accuracy": 0.1277147874236107,
|
|
"num_tokens": 3285817.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.372090244293213,
|
|
"epoch": 1.41988459831336,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004962260740895398,
|
|
"loss": 6.2855,
|
|
"mean_token_accuracy": 0.13827268779277802,
|
|
"num_tokens": 3296303.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.379140472412109,
|
|
"epoch": 1.424323124722592,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004961629877939913,
|
|
"loss": 6.3871,
|
|
"mean_token_accuracy": 0.1381452649831772,
|
|
"num_tokens": 3305686.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.410601949691772,
|
|
"epoch": 1.4287616511318242,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496099383107511,
|
|
"loss": 6.3337,
|
|
"mean_token_accuracy": 0.13385034054517747,
|
|
"num_tokens": 3316243.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.363243675231933,
|
|
"epoch": 1.4332001775410563,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004960352601791835,
|
|
"loss": 6.3509,
|
|
"mean_token_accuracy": 0.1326597101986408,
|
|
"num_tokens": 3326118.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.420337390899658,
|
|
"epoch": 1.4376387039502885,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004959706191593087,
|
|
"loss": 6.2826,
|
|
"mean_token_accuracy": 0.139851226657629,
|
|
"num_tokens": 3336597.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.367270183563233,
|
|
"epoch": 1.4420772303595206,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004959054601994007,
|
|
"loss": 6.3475,
|
|
"mean_token_accuracy": 0.13378802165389062,
|
|
"num_tokens": 3346958.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.393503141403198,
|
|
"epoch": 1.4465157567687528,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004958397834521878,
|
|
"loss": 6.3568,
|
|
"mean_token_accuracy": 0.1363594651222229,
|
|
"num_tokens": 3357068.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.357707118988037,
|
|
"epoch": 1.450954283177985,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004957735890716115,
|
|
"loss": 6.3272,
|
|
"mean_token_accuracy": 0.1293055720627308,
|
|
"num_tokens": 3367710.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.438919878005981,
|
|
"epoch": 1.455392809587217,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004957068772128273,
|
|
"loss": 6.3926,
|
|
"mean_token_accuracy": 0.12550436183810235,
|
|
"num_tokens": 3377146.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.4241721630096436,
|
|
"epoch": 1.4598313359964492,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000495639648032203,
|
|
"loss": 6.2687,
|
|
"mean_token_accuracy": 0.13697705715894698,
|
|
"num_tokens": 3387494.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.404942893981934,
|
|
"epoch": 1.4642698624056814,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004955719016873192,
|
|
"loss": 6.271,
|
|
"mean_token_accuracy": 0.1324605606496334,
|
|
"num_tokens": 3397667.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.341697072982788,
|
|
"epoch": 1.4687083888149135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004955036383369688,
|
|
"loss": 6.2882,
|
|
"mean_token_accuracy": 0.13828188329935073,
|
|
"num_tokens": 3407245.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.33912615776062,
|
|
"epoch": 1.4731469152241456,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004954348581411564,
|
|
"loss": 6.3502,
|
|
"mean_token_accuracy": 0.13207383826375008,
|
|
"num_tokens": 3417796.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.365179395675659,
|
|
"epoch": 1.4775854416333778,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000495365561261098,
|
|
"loss": 6.3521,
|
|
"mean_token_accuracy": 0.13066228553652764,
|
|
"num_tokens": 3427665.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.460514163970947,
|
|
"epoch": 1.48202396804261,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004952957478592209,
|
|
"loss": 6.4161,
|
|
"mean_token_accuracy": 0.13068753182888032,
|
|
"num_tokens": 3438279.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.346782398223877,
|
|
"epoch": 1.486462494451842,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004952254180991628,
|
|
"loss": 6.3924,
|
|
"mean_token_accuracy": 0.1330470062792301,
|
|
"num_tokens": 3448722.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.355397272109985,
|
|
"epoch": 1.4909010208610742,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004951545721457719,
|
|
"loss": 6.3045,
|
|
"mean_token_accuracy": 0.14068271815776826,
|
|
"num_tokens": 3459050.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.3841273307800295,
|
|
"epoch": 1.4953395472703064,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 6.2644,
|
|
"mean_token_accuracy": 0.13605008125305176,
|
|
"num_tokens": 3468833.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.393172645568848,
|
|
"epoch": 1.4997780736795385,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004950113323244336,
|
|
"loss": 6.3353,
|
|
"mean_token_accuracy": 0.13272473365068435,
|
|
"num_tokens": 3479119.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.422500705718994,
|
|
"epoch": 1.5042166000887707,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004949389387922305,
|
|
"loss": 6.3001,
|
|
"mean_token_accuracy": 0.13518199026584626,
|
|
"num_tokens": 3490339.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.375849771499634,
|
|
"epoch": 1.5086551264980028,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004948660297381826,
|
|
"loss": 6.2959,
|
|
"mean_token_accuracy": 0.13820037841796876,
|
|
"num_tokens": 3500645.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.326462697982788,
|
|
"epoch": 1.5130936529072347,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004947926053331836,
|
|
"loss": 6.273,
|
|
"mean_token_accuracy": 0.139491256326437,
|
|
"num_tokens": 3510393.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.408394193649292,
|
|
"epoch": 1.517532179316467,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004947186657493354,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.14247287064790726,
|
|
"num_tokens": 3520707.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.333920478820801,
|
|
"epoch": 1.521970705725699,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004946442111599473,
|
|
"loss": 6.4165,
|
|
"mean_token_accuracy": 0.13150266110897063,
|
|
"num_tokens": 3532101.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.405943059921265,
|
|
"epoch": 1.5264092321349312,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004945692417395358,
|
|
"loss": 6.2607,
|
|
"mean_token_accuracy": 0.13934548795223237,
|
|
"num_tokens": 3541270.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.301887083053589,
|
|
"epoch": 1.5308477585441633,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000494493757663824,
|
|
"loss": 6.3411,
|
|
"mean_token_accuracy": 0.13587306886911393,
|
|
"num_tokens": 3550959.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.384119939804077,
|
|
"epoch": 1.5352862849533955,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004944177591097415,
|
|
"loss": 6.3751,
|
|
"mean_token_accuracy": 0.13755866140127182,
|
|
"num_tokens": 3561853.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.344173955917358,
|
|
"epoch": 1.5397248113626276,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004943412462554236,
|
|
"loss": 6.3123,
|
|
"mean_token_accuracy": 0.13452489599585532,
|
|
"num_tokens": 3573060.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.308044004440307,
|
|
"epoch": 1.5441633377718598,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004942642192802114,
|
|
"loss": 6.328,
|
|
"mean_token_accuracy": 0.1374310664832592,
|
|
"num_tokens": 3582777.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.439042711257935,
|
|
"epoch": 1.548601864181092,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004941866783646508,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.13529076278209687,
|
|
"num_tokens": 3592978.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.348781871795654,
|
|
"epoch": 1.553040390590324,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004941086236904923,
|
|
"loss": 6.3061,
|
|
"mean_token_accuracy": 0.13936796635389329,
|
|
"num_tokens": 3603355.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.436870050430298,
|
|
"epoch": 1.557478916999556,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004940300554406909,
|
|
"loss": 6.3498,
|
|
"mean_token_accuracy": 0.1272510677576065,
|
|
"num_tokens": 3614843.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.3685750484466555,
|
|
"epoch": 1.5619174434087881,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000493950973799405,
|
|
"loss": 6.2738,
|
|
"mean_token_accuracy": 0.14151667505502702,
|
|
"num_tokens": 3623953.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.366236734390259,
|
|
"epoch": 1.5663559698180203,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004938713789519967,
|
|
"loss": 6.3528,
|
|
"mean_token_accuracy": 0.13301268741488456,
|
|
"num_tokens": 3634762.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.44322018623352,
|
|
"epoch": 1.5707944962272524,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004937912710850309,
|
|
"loss": 6.3909,
|
|
"mean_token_accuracy": 0.12702670469880104,
|
|
"num_tokens": 3645597.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.3498796939849855,
|
|
"epoch": 1.5752330226364846,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004937106503862749,
|
|
"loss": 6.2962,
|
|
"mean_token_accuracy": 0.1402463473379612,
|
|
"num_tokens": 3655895.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.3070460796356205,
|
|
"epoch": 1.5796715490457167,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004936295170446981,
|
|
"loss": 6.2705,
|
|
"mean_token_accuracy": 0.13688953965902328,
|
|
"num_tokens": 3667181.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.274955320358276,
|
|
"epoch": 1.5841100754549489,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004935478712504715,
|
|
"loss": 6.2313,
|
|
"mean_token_accuracy": 0.14355697929859162,
|
|
"num_tokens": 3676675.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.3787109375,
|
|
"epoch": 1.588548601864181,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004934657131949674,
|
|
"loss": 6.293,
|
|
"mean_token_accuracy": 0.13846610561013223,
|
|
"num_tokens": 3687354.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.278066110610962,
|
|
"epoch": 1.5929871282734132,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004933830430707585,
|
|
"loss": 6.3331,
|
|
"mean_token_accuracy": 0.1389099143445492,
|
|
"num_tokens": 3696762.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.403505325317383,
|
|
"epoch": 1.5974256546826453,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000493299861071618,
|
|
"loss": 6.3193,
|
|
"mean_token_accuracy": 0.13765995875000953,
|
|
"num_tokens": 3706671.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.346765232086182,
|
|
"epoch": 1.6018641810918774,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004932161673925189,
|
|
"loss": 6.3211,
|
|
"mean_token_accuracy": 0.14222623705863952,
|
|
"num_tokens": 3716047.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.480462837219238,
|
|
"epoch": 1.6063027075011096,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004931319622296333,
|
|
"loss": 6.3733,
|
|
"mean_token_accuracy": 0.12890343070030214,
|
|
"num_tokens": 3728249.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.327421188354492,
|
|
"epoch": 1.6107412339103417,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004930472457803324,
|
|
"loss": 6.2919,
|
|
"mean_token_accuracy": 0.14020831808447837,
|
|
"num_tokens": 3739164.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.408966207504273,
|
|
"epoch": 1.6151797603195739,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004929620182431858,
|
|
"loss": 6.303,
|
|
"mean_token_accuracy": 0.142412880808115,
|
|
"num_tokens": 3749190.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.261825037002564,
|
|
"epoch": 1.619618286728806,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004928762798179612,
|
|
"loss": 6.2169,
|
|
"mean_token_accuracy": 0.1475951187312603,
|
|
"num_tokens": 3760265.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.274430656433106,
|
|
"epoch": 1.6240568131380382,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004927900307056233,
|
|
"loss": 6.2945,
|
|
"mean_token_accuracy": 0.13552614152431489,
|
|
"num_tokens": 3771068.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.3409483432769775,
|
|
"epoch": 1.6284953395472703,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004927032711083342,
|
|
"loss": 6.297,
|
|
"mean_token_accuracy": 0.13993098884820937,
|
|
"num_tokens": 3781230.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.295605039596557,
|
|
"epoch": 1.6329338659565025,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004926160012294526,
|
|
"loss": 6.2502,
|
|
"mean_token_accuracy": 0.14423105642199516,
|
|
"num_tokens": 3791435.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.265171527862549,
|
|
"epoch": 1.6373723923657346,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000492528221273533,
|
|
"loss": 6.2049,
|
|
"mean_token_accuracy": 0.14195780605077743,
|
|
"num_tokens": 3801680.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.373475074768066,
|
|
"epoch": 1.6418109187749668,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004924399314463258,
|
|
"loss": 6.3058,
|
|
"mean_token_accuracy": 0.13744521513581276,
|
|
"num_tokens": 3812005.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.294007301330566,
|
|
"epoch": 1.646249445184199,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004923511319547761,
|
|
"loss": 6.2355,
|
|
"mean_token_accuracy": 0.1440819539129734,
|
|
"num_tokens": 3821682.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.319647216796875,
|
|
"epoch": 1.650687971593431,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000492261823007024,
|
|
"loss": 6.2157,
|
|
"mean_token_accuracy": 0.14868459478020668,
|
|
"num_tokens": 3831725.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.311188554763794,
|
|
"epoch": 1.6551264980026632,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004921720048124034,
|
|
"loss": 6.2569,
|
|
"mean_token_accuracy": 0.13653967306017875,
|
|
"num_tokens": 3840647.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.402250814437866,
|
|
"epoch": 1.6595650244118954,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004920816775814422,
|
|
"loss": 6.3124,
|
|
"mean_token_accuracy": 0.1325970098376274,
|
|
"num_tokens": 3852869.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.3165970802307125,
|
|
"epoch": 1.6640035508211275,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 6.2912,
|
|
"mean_token_accuracy": 0.14089620634913444,
|
|
"num_tokens": 3863484.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.288156318664551,
|
|
"epoch": 1.6684420772303596,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000491899496858574,
|
|
"loss": 6.2841,
|
|
"mean_token_accuracy": 0.13467081785202026,
|
|
"num_tokens": 3874564.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.395035028457642,
|
|
"epoch": 1.6728806036395918,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004918076437936859,
|
|
"loss": 6.3119,
|
|
"mean_token_accuracy": 0.1333139009773731,
|
|
"num_tokens": 3884315.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.327268505096436,
|
|
"epoch": 1.677319130048824,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004917152825464947,
|
|
"loss": 6.2965,
|
|
"mean_token_accuracy": 0.13294868022203446,
|
|
"num_tokens": 3893948.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.396142435073853,
|
|
"epoch": 1.681757656458056,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004916224133334885,
|
|
"loss": 6.311,
|
|
"mean_token_accuracy": 0.13014644309878348,
|
|
"num_tokens": 3905099.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.286337661743164,
|
|
"epoch": 1.686196182867288,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004915290363723465,
|
|
"loss": 6.2342,
|
|
"mean_token_accuracy": 0.14227957800030708,
|
|
"num_tokens": 3914674.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.324175691604614,
|
|
"epoch": 1.6906347092765202,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004914351518819379,
|
|
"loss": 6.2272,
|
|
"mean_token_accuracy": 0.14022424072027206,
|
|
"num_tokens": 3925096.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.305720806121826,
|
|
"epoch": 1.6950732356857523,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004913407600823216,
|
|
"loss": 6.2887,
|
|
"mean_token_accuracy": 0.1422324001789093,
|
|
"num_tokens": 3935048.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.279489135742187,
|
|
"epoch": 1.6995117620949844,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004912458611947454,
|
|
"loss": 6.2333,
|
|
"mean_token_accuracy": 0.14147737249732018,
|
|
"num_tokens": 3944021.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.27567949295044,
|
|
"epoch": 1.7039502885042166,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000491150455441646,
|
|
"loss": 6.166,
|
|
"mean_token_accuracy": 0.14739958047866822,
|
|
"num_tokens": 3953587.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.247754430770874,
|
|
"epoch": 1.7083888149134487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004910545430466478,
|
|
"loss": 6.2706,
|
|
"mean_token_accuracy": 0.13362218514084817,
|
|
"num_tokens": 3963981.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.270370864868164,
|
|
"epoch": 1.7128273413226809,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004909581242345628,
|
|
"loss": 6.2833,
|
|
"mean_token_accuracy": 0.14019264951348304,
|
|
"num_tokens": 3973720.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.295551729202271,
|
|
"epoch": 1.717265867731913,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004908611992313905,
|
|
"loss": 6.222,
|
|
"mean_token_accuracy": 0.14261915981769563,
|
|
"num_tokens": 3984579.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.264960384368896,
|
|
"epoch": 1.7217043941411452,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004907637682643162,
|
|
"loss": 6.245,
|
|
"mean_token_accuracy": 0.1404101848602295,
|
|
"num_tokens": 3994609.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.302646636962891,
|
|
"epoch": 1.7261429205503773,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004906658315617112,
|
|
"loss": 6.2902,
|
|
"mean_token_accuracy": 0.13819315508008004,
|
|
"num_tokens": 4004969.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.248818016052246,
|
|
"epoch": 1.7305814469596092,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004905673893531331,
|
|
"loss": 6.3268,
|
|
"mean_token_accuracy": 0.1326567329466343,
|
|
"num_tokens": 4016098.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.397868633270264,
|
|
"epoch": 1.7350199733688414,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004904684418693231,
|
|
"loss": 6.233,
|
|
"mean_token_accuracy": 0.14564677476882934,
|
|
"num_tokens": 4025744.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.238390398025513,
|
|
"epoch": 1.7394584997780735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004903689893422077,
|
|
"loss": 6.2537,
|
|
"mean_token_accuracy": 0.13860218822956086,
|
|
"num_tokens": 4036532.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.4320872783660885,
|
|
"epoch": 1.7438970261873057,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004902690320048966,
|
|
"loss": 6.3386,
|
|
"mean_token_accuracy": 0.13317947015166282,
|
|
"num_tokens": 4047350.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.341959428787232,
|
|
"epoch": 1.7483355525965378,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000490168570091683,
|
|
"loss": 6.3134,
|
|
"mean_token_accuracy": 0.14479754120111465,
|
|
"num_tokens": 4058223.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.26735486984253,
|
|
"epoch": 1.75277407900577,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004900676038380429,
|
|
"loss": 6.2377,
|
|
"mean_token_accuracy": 0.1400671385228634,
|
|
"num_tokens": 4067741.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.2349577903747555,
|
|
"epoch": 1.7572126054150021,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004899661334806342,
|
|
"loss": 6.2507,
|
|
"mean_token_accuracy": 0.1395649529993534,
|
|
"num_tokens": 4078598.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.328679370880127,
|
|
"epoch": 1.7616511318242343,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004898641592572965,
|
|
"loss": 6.2545,
|
|
"mean_token_accuracy": 0.1380497135221958,
|
|
"num_tokens": 4089230.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.289399909973144,
|
|
"epoch": 1.7660896582334664,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004897616814070505,
|
|
"loss": 6.1316,
|
|
"mean_token_accuracy": 0.14506224393844605,
|
|
"num_tokens": 4099104.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.196705055236817,
|
|
"epoch": 1.7705281846426986,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004896587001700972,
|
|
"loss": 6.2359,
|
|
"mean_token_accuracy": 0.13811009451746942,
|
|
"num_tokens": 4109008.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.3165913105010985,
|
|
"epoch": 1.7749667110519307,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895552157878174,
|
|
"loss": 6.1727,
|
|
"mean_token_accuracy": 0.14556334167718887,
|
|
"num_tokens": 4118219.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7749667110519307,
|
|
"eval_entropy": 6.189312974517533,
|
|
"eval_loss": 6.335569858551025,
|
|
"eval_mean_token_accuracy": 0.136928480658282,
|
|
"eval_num_tokens": 4118219.0,
|
|
"eval_runtime": 2.7093,
|
|
"eval_samples_per_second": 1242.755,
|
|
"eval_steps_per_second": 155.39,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11260,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 6244425365760000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|