ec2fe396ab
Model: fpadovani/tur-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
11156 lines
304 KiB
JSON
11156 lines
304 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 5.467429139731477,
|
|
"eval_steps": 500,
|
|
"global_step": 5500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.810139894485474,
|
|
"epoch": 0.004972650422675286,
|
|
"grad_norm": 21.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.2169,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9089.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.796057510375976,
|
|
"epoch": 0.009945300845350571,
|
|
"grad_norm": 22.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.3796,
|
|
"mean_token_accuracy": 0.0001728883828036487,
|
|
"num_tokens": 21650.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.851445960998535,
|
|
"epoch": 0.014917951268025857,
|
|
"grad_norm": 24.75,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.047,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 31067.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.946325588226318,
|
|
"epoch": 0.019890601690701143,
|
|
"grad_norm": 32.0,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.8176,
|
|
"mean_token_accuracy": 0.0005868025938980282,
|
|
"num_tokens": 40816.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.410843563079834,
|
|
"epoch": 0.02486325211337643,
|
|
"grad_norm": 40.25,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.0547,
|
|
"mean_token_accuracy": 0.0003202892781700939,
|
|
"num_tokens": 51065.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.1249548435211185,
|
|
"epoch": 0.029835902536051714,
|
|
"grad_norm": 30.125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.8315,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 60539.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.19781551361084,
|
|
"epoch": 0.034808552958727,
|
|
"grad_norm": 3.3125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.8532,
|
|
"mean_token_accuracy": 0.0021006516821216793,
|
|
"num_tokens": 71010.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.736362266540528,
|
|
"epoch": 0.039781203381402286,
|
|
"grad_norm": 3.125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.6601,
|
|
"mean_token_accuracy": 0.012861686293035746,
|
|
"num_tokens": 81394.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.72396240234375,
|
|
"epoch": 0.04475385380407757,
|
|
"grad_norm": 2.875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.5728,
|
|
"mean_token_accuracy": 0.015258754882961512,
|
|
"num_tokens": 92491.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.611907863616944,
|
|
"epoch": 0.04972650422675286,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.3994,
|
|
"mean_token_accuracy": 0.01761980978772044,
|
|
"num_tokens": 102113.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.650904369354247,
|
|
"epoch": 0.05469915464942814,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.3041,
|
|
"mean_token_accuracy": 0.01719568707048893,
|
|
"num_tokens": 112211.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.714541339874268,
|
|
"epoch": 0.05967180507210343,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1449,
|
|
"mean_token_accuracy": 0.02930256463587284,
|
|
"num_tokens": 122166.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.710609722137452,
|
|
"epoch": 0.06464445549477872,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0419,
|
|
"mean_token_accuracy": 0.04249542821198702,
|
|
"num_tokens": 131016.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.672384357452392,
|
|
"epoch": 0.069617105917454,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.9842,
|
|
"mean_token_accuracy": 0.03610398881137371,
|
|
"num_tokens": 141369.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.683287620544434,
|
|
"epoch": 0.07458975634012929,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.9433,
|
|
"mean_token_accuracy": 0.051322261616587636,
|
|
"num_tokens": 150939.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.683112144470215,
|
|
"epoch": 0.07956240676280457,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.9042,
|
|
"mean_token_accuracy": 0.050704098865389825,
|
|
"num_tokens": 159981.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.666728496551514,
|
|
"epoch": 0.08453505718547986,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.8301,
|
|
"mean_token_accuracy": 0.0480174720287323,
|
|
"num_tokens": 170139.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.6644793510437,
|
|
"epoch": 0.08950770760815514,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.767,
|
|
"mean_token_accuracy": 0.05141008589416742,
|
|
"num_tokens": 180746.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.65702543258667,
|
|
"epoch": 0.09448035803083044,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.6469,
|
|
"mean_token_accuracy": 0.057331408560276034,
|
|
"num_tokens": 190250.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.622721958160401,
|
|
"epoch": 0.09945300845350571,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.6532,
|
|
"mean_token_accuracy": 0.04415315557271242,
|
|
"num_tokens": 201185.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.592253017425538,
|
|
"epoch": 0.10442565887618101,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.5616,
|
|
"mean_token_accuracy": 0.05255695134401321,
|
|
"num_tokens": 211476.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.56272497177124,
|
|
"epoch": 0.10939830929885629,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.4667,
|
|
"mean_token_accuracy": 0.05462103560566902,
|
|
"num_tokens": 220965.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.503788185119628,
|
|
"epoch": 0.11437095972153158,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.4022,
|
|
"mean_token_accuracy": 0.04950843863189221,
|
|
"num_tokens": 231539.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.478209018707275,
|
|
"epoch": 0.11934361014420686,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.3051,
|
|
"mean_token_accuracy": 0.05306029766798019,
|
|
"num_tokens": 241206.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.38660650253296,
|
|
"epoch": 0.12431626056688215,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.2202,
|
|
"mean_token_accuracy": 0.05398030839860439,
|
|
"num_tokens": 250629.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.396561527252198,
|
|
"epoch": 0.12928891098955744,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.1838,
|
|
"mean_token_accuracy": 0.0500319853425026,
|
|
"num_tokens": 260705.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.283761501312256,
|
|
"epoch": 0.13426156141223272,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.1281,
|
|
"mean_token_accuracy": 0.0536313846707344,
|
|
"num_tokens": 270499.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.26081371307373,
|
|
"epoch": 0.139234211834908,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 9.0683,
|
|
"mean_token_accuracy": 0.05152654591947794,
|
|
"num_tokens": 279998.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.157192802429199,
|
|
"epoch": 0.14420686225758328,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.9399,
|
|
"mean_token_accuracy": 0.05353470556437969,
|
|
"num_tokens": 290322.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.07140588760376,
|
|
"epoch": 0.14917951268025859,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.858,
|
|
"mean_token_accuracy": 0.059231129661202434,
|
|
"num_tokens": 299893.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.94995050430298,
|
|
"epoch": 0.15415216310293386,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.7989,
|
|
"mean_token_accuracy": 0.05670776516199112,
|
|
"num_tokens": 309191.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.815523719787597,
|
|
"epoch": 0.15912481352560914,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.7036,
|
|
"mean_token_accuracy": 0.057601967453956605,
|
|
"num_tokens": 318494.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.753715515136719,
|
|
"epoch": 0.16409746394828442,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.6798,
|
|
"mean_token_accuracy": 0.05443971864879131,
|
|
"num_tokens": 328299.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.610096073150634,
|
|
"epoch": 0.16907011437095973,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.6986,
|
|
"mean_token_accuracy": 0.05396176092326641,
|
|
"num_tokens": 338878.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.495657157897949,
|
|
"epoch": 0.174042764793635,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.5977,
|
|
"mean_token_accuracy": 0.05288307964801788,
|
|
"num_tokens": 348508.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.364479351043702,
|
|
"epoch": 0.1790154152163103,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.5384,
|
|
"mean_token_accuracy": 0.060325421020388605,
|
|
"num_tokens": 358233.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.208246898651122,
|
|
"epoch": 0.1839880656389856,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.4774,
|
|
"mean_token_accuracy": 0.05886298380792141,
|
|
"num_tokens": 367297.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.103068447113037,
|
|
"epoch": 0.18896071606166087,
|
|
"grad_norm": 0.671875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.471,
|
|
"mean_token_accuracy": 0.052080392837524414,
|
|
"num_tokens": 376880.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 9.05744981765747,
|
|
"epoch": 0.19393336648433615,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.4617,
|
|
"mean_token_accuracy": 0.05685535296797752,
|
|
"num_tokens": 386806.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.958063793182372,
|
|
"epoch": 0.19890601690701143,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.4048,
|
|
"mean_token_accuracy": 0.059261714294552806,
|
|
"num_tokens": 396161.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.898060321807861,
|
|
"epoch": 0.20387866732968674,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.4628,
|
|
"mean_token_accuracy": 0.057381686940789225,
|
|
"num_tokens": 405643.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.85878095626831,
|
|
"epoch": 0.20885131775236201,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.4071,
|
|
"mean_token_accuracy": 0.0663915179669857,
|
|
"num_tokens": 414367.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.785406398773194,
|
|
"epoch": 0.2138239681750373,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.4544,
|
|
"mean_token_accuracy": 0.05882962830364704,
|
|
"num_tokens": 424430.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.815994358062744,
|
|
"epoch": 0.21879661859771257,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.4285,
|
|
"mean_token_accuracy": 0.06256655529141426,
|
|
"num_tokens": 435176.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.771424484252929,
|
|
"epoch": 0.22376926902038788,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.4408,
|
|
"mean_token_accuracy": 0.060541344434022905,
|
|
"num_tokens": 445611.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.745358943939209,
|
|
"epoch": 0.22874191944306316,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.3751,
|
|
"mean_token_accuracy": 0.0634678304195404,
|
|
"num_tokens": 455143.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.766716575622558,
|
|
"epoch": 0.23371456986573844,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.4738,
|
|
"mean_token_accuracy": 0.06072634160518646,
|
|
"num_tokens": 465507.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.770550727844238,
|
|
"epoch": 0.23868722028841372,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.4583,
|
|
"mean_token_accuracy": 0.06104334816336632,
|
|
"num_tokens": 476152.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.77504940032959,
|
|
"epoch": 0.24365987071108902,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.3132,
|
|
"mean_token_accuracy": 0.06499341167509556,
|
|
"num_tokens": 486341.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.655905055999757,
|
|
"epoch": 0.2486325211337643,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.2634,
|
|
"mean_token_accuracy": 0.06762947700917721,
|
|
"num_tokens": 495915.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.659723567962647,
|
|
"epoch": 0.2536051715564396,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.3514,
|
|
"mean_token_accuracy": 0.06988421343266964,
|
|
"num_tokens": 505666.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.692667961120605,
|
|
"epoch": 0.2585778219791149,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.3023,
|
|
"mean_token_accuracy": 0.0659954596310854,
|
|
"num_tokens": 515411.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.63499994277954,
|
|
"epoch": 0.26355047240179014,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.3494,
|
|
"mean_token_accuracy": 0.06087326742708683,
|
|
"num_tokens": 525018.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.6294114112854,
|
|
"epoch": 0.26852312282446544,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.3103,
|
|
"mean_token_accuracy": 0.06277070567011833,
|
|
"num_tokens": 535056.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.69818925857544,
|
|
"epoch": 0.27349577324714075,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.309,
|
|
"mean_token_accuracy": 0.06765819452703,
|
|
"num_tokens": 545115.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.642760372161865,
|
|
"epoch": 0.278468423669816,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.3213,
|
|
"mean_token_accuracy": 0.06374539136886596,
|
|
"num_tokens": 554429.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.534232425689698,
|
|
"epoch": 0.2834410740924913,
|
|
"grad_norm": 0.6640625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2617,
|
|
"mean_token_accuracy": 0.0756473757326603,
|
|
"num_tokens": 563357.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.696459770202637,
|
|
"epoch": 0.28841372451516656,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.3499,
|
|
"mean_token_accuracy": 0.0666810642927885,
|
|
"num_tokens": 574480.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.571897983551025,
|
|
"epoch": 0.29338637493784187,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.2731,
|
|
"mean_token_accuracy": 0.06942101046442986,
|
|
"num_tokens": 583722.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.63041067123413,
|
|
"epoch": 0.29835902536051717,
|
|
"grad_norm": 0.66015625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.2406,
|
|
"mean_token_accuracy": 0.06790350154042243,
|
|
"num_tokens": 593644.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.592833423614502,
|
|
"epoch": 0.3033316757831924,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.2753,
|
|
"mean_token_accuracy": 0.06820234544575214,
|
|
"num_tokens": 603547.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.531950759887696,
|
|
"epoch": 0.30830432620586773,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.3059,
|
|
"mean_token_accuracy": 0.06576361991465092,
|
|
"num_tokens": 613880.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.58628578186035,
|
|
"epoch": 0.31327697662854304,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.288,
|
|
"mean_token_accuracy": 0.06338743269443511,
|
|
"num_tokens": 624213.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.631004810333252,
|
|
"epoch": 0.3182496270512183,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.3069,
|
|
"mean_token_accuracy": 0.06901536360383034,
|
|
"num_tokens": 634350.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.613157272338867,
|
|
"epoch": 0.3232222774738936,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.3072,
|
|
"mean_token_accuracy": 0.06461950577795506,
|
|
"num_tokens": 644040.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.611722564697265,
|
|
"epoch": 0.32819492789656884,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.2689,
|
|
"mean_token_accuracy": 0.070689857006073,
|
|
"num_tokens": 652915.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.554406261444091,
|
|
"epoch": 0.33316757831924415,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.2508,
|
|
"mean_token_accuracy": 0.06666112951934337,
|
|
"num_tokens": 662665.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.609964942932129,
|
|
"epoch": 0.33814022874191946,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.2467,
|
|
"mean_token_accuracy": 0.07032966800034046,
|
|
"num_tokens": 671667.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.529174995422363,
|
|
"epoch": 0.3431128791645947,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.2799,
|
|
"mean_token_accuracy": 0.06516398154199124,
|
|
"num_tokens": 680432.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.62266035079956,
|
|
"epoch": 0.34808552958727,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.2702,
|
|
"mean_token_accuracy": 0.06741996183991432,
|
|
"num_tokens": 690184.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.603531074523925,
|
|
"epoch": 0.3530581800099453,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.159,
|
|
"mean_token_accuracy": 0.07316870428621769,
|
|
"num_tokens": 699147.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.518417072296142,
|
|
"epoch": 0.3580308304326206,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1926,
|
|
"mean_token_accuracy": 0.0727127481251955,
|
|
"num_tokens": 708696.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.589232635498046,
|
|
"epoch": 0.3630034808552959,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.1562,
|
|
"mean_token_accuracy": 0.06915070340037346,
|
|
"num_tokens": 718438.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.530152130126954,
|
|
"epoch": 0.3679761312779712,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.2959,
|
|
"mean_token_accuracy": 0.06483390927314758,
|
|
"num_tokens": 728157.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.543238544464112,
|
|
"epoch": 0.37294878170064644,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.1901,
|
|
"mean_token_accuracy": 0.07307061739265919,
|
|
"num_tokens": 737007.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.549563598632812,
|
|
"epoch": 0.37792143212332174,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.1671,
|
|
"mean_token_accuracy": 0.07229796946048736,
|
|
"num_tokens": 746873.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.522933387756348,
|
|
"epoch": 0.382894082545997,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.0514,
|
|
"mean_token_accuracy": 0.08149674609303474,
|
|
"num_tokens": 755486.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.475645160675048,
|
|
"epoch": 0.3878667329686723,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.1482,
|
|
"mean_token_accuracy": 0.07295819409191609,
|
|
"num_tokens": 765481.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.463741111755372,
|
|
"epoch": 0.3928393833913476,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.1135,
|
|
"mean_token_accuracy": 0.07425905987620354,
|
|
"num_tokens": 774335.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.596048164367676,
|
|
"epoch": 0.39781203381402286,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.2392,
|
|
"mean_token_accuracy": 0.07124232538044453,
|
|
"num_tokens": 784307.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.549865531921387,
|
|
"epoch": 0.40278468423669817,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.2106,
|
|
"mean_token_accuracy": 0.07055418565869331,
|
|
"num_tokens": 795058.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.556127262115478,
|
|
"epoch": 0.40775733465937347,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.1755,
|
|
"mean_token_accuracy": 0.07405509985983372,
|
|
"num_tokens": 805687.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.495815467834472,
|
|
"epoch": 0.4127299850820487,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.1541,
|
|
"mean_token_accuracy": 0.07714710012078285,
|
|
"num_tokens": 815138.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.47137746810913,
|
|
"epoch": 0.41770263550472403,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.1172,
|
|
"mean_token_accuracy": 0.07492221929132939,
|
|
"num_tokens": 825009.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.46223554611206,
|
|
"epoch": 0.4226752859273993,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.1004,
|
|
"mean_token_accuracy": 0.07772790230810642,
|
|
"num_tokens": 834708.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.475219249725342,
|
|
"epoch": 0.4276479363500746,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.1753,
|
|
"mean_token_accuracy": 0.07542566247284413,
|
|
"num_tokens": 843888.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.477521610260009,
|
|
"epoch": 0.4326205867727499,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.1222,
|
|
"mean_token_accuracy": 0.07436901330947876,
|
|
"num_tokens": 853057.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.48756446838379,
|
|
"epoch": 0.43759323719542514,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.1761,
|
|
"mean_token_accuracy": 0.07114730812609196,
|
|
"num_tokens": 862701.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.38951072692871,
|
|
"epoch": 0.44256588761810045,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0672,
|
|
"mean_token_accuracy": 0.07624398544430733,
|
|
"num_tokens": 872129.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.414244651794434,
|
|
"epoch": 0.44753853804077576,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.0759,
|
|
"mean_token_accuracy": 0.0760790079832077,
|
|
"num_tokens": 883166.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.476319980621337,
|
|
"epoch": 0.452511188463451,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.0367,
|
|
"mean_token_accuracy": 0.08530011028051376,
|
|
"num_tokens": 894159.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.479471683502197,
|
|
"epoch": 0.4574838388861263,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0962,
|
|
"mean_token_accuracy": 0.07649504765868187,
|
|
"num_tokens": 904588.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.296580982208251,
|
|
"epoch": 0.46245648930880157,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.0798,
|
|
"mean_token_accuracy": 0.0781316339969635,
|
|
"num_tokens": 915554.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.486116504669189,
|
|
"epoch": 0.4674291397314769,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.0891,
|
|
"mean_token_accuracy": 0.07881890460848809,
|
|
"num_tokens": 925546.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.337780475616455,
|
|
"epoch": 0.4724017901541522,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.0598,
|
|
"mean_token_accuracy": 0.07934539914131164,
|
|
"num_tokens": 936615.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.46543378829956,
|
|
"epoch": 0.47737444057682743,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.0858,
|
|
"mean_token_accuracy": 0.07611360922455787,
|
|
"num_tokens": 947374.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.334126091003418,
|
|
"epoch": 0.48234709099950274,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.0012,
|
|
"mean_token_accuracy": 0.08066745214164257,
|
|
"num_tokens": 957772.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.434980773925782,
|
|
"epoch": 0.48731974142217804,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.0842,
|
|
"mean_token_accuracy": 0.07925943285226822,
|
|
"num_tokens": 967724.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.30348539352417,
|
|
"epoch": 0.4922923918448533,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9835,
|
|
"mean_token_accuracy": 0.07751845642924308,
|
|
"num_tokens": 976497.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.317158412933349,
|
|
"epoch": 0.4972650422675286,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9707,
|
|
"mean_token_accuracy": 0.08364193737506867,
|
|
"num_tokens": 986207.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4972650422675286,
|
|
"eval_entropy": 8.216110460219845,
|
|
"eval_loss": 8.090629577636719,
|
|
"eval_mean_token_accuracy": 0.07677195617748846,
|
|
"eval_num_tokens": 986207.0,
|
|
"eval_runtime": 1.9669,
|
|
"eval_samples_per_second": 1761.677,
|
|
"eval_steps_per_second": 220.654,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.281911659240723,
|
|
"epoch": 0.5022376926902039,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.9324,
|
|
"mean_token_accuracy": 0.08303811699151993,
|
|
"num_tokens": 995568.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.333969402313233,
|
|
"epoch": 0.5072103431128792,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 8.0045,
|
|
"mean_token_accuracy": 0.0778925932943821,
|
|
"num_tokens": 1005829.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.28199052810669,
|
|
"epoch": 0.5121829935355544,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.9612,
|
|
"mean_token_accuracy": 0.07643497064709663,
|
|
"num_tokens": 1015290.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.325776958465577,
|
|
"epoch": 0.5171556439582298,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9266,
|
|
"mean_token_accuracy": 0.08521686196327209,
|
|
"num_tokens": 1024479.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.214553546905517,
|
|
"epoch": 0.522128294380905,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.8858,
|
|
"mean_token_accuracy": 0.08373716920614242,
|
|
"num_tokens": 1034092.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.403457546234131,
|
|
"epoch": 0.5271009448035803,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 8.0627,
|
|
"mean_token_accuracy": 0.08292205855250359,
|
|
"num_tokens": 1043844.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.235153484344483,
|
|
"epoch": 0.5320735952262556,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.9054,
|
|
"mean_token_accuracy": 0.08438774272799492,
|
|
"num_tokens": 1052646.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.292011642456055,
|
|
"epoch": 0.5370462456489309,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.9586,
|
|
"mean_token_accuracy": 0.08071774989366531,
|
|
"num_tokens": 1062345.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.369629669189454,
|
|
"epoch": 0.5420188960716061,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.946,
|
|
"mean_token_accuracy": 0.08147362023591995,
|
|
"num_tokens": 1071116.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.257204055786133,
|
|
"epoch": 0.5469915464942815,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.9046,
|
|
"mean_token_accuracy": 0.08467757701873779,
|
|
"num_tokens": 1080562.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.170803356170655,
|
|
"epoch": 0.5519641969169568,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.966,
|
|
"mean_token_accuracy": 0.0811925157904625,
|
|
"num_tokens": 1091944.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.311559867858886,
|
|
"epoch": 0.556936847339632,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8989,
|
|
"mean_token_accuracy": 0.08553457483649254,
|
|
"num_tokens": 1101412.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.295998668670654,
|
|
"epoch": 0.5619094977623074,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.964,
|
|
"mean_token_accuracy": 0.08476055823266507,
|
|
"num_tokens": 1110652.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.243193197250367,
|
|
"epoch": 0.5668821481849826,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8749,
|
|
"mean_token_accuracy": 0.08798420876264572,
|
|
"num_tokens": 1120420.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.29685411453247,
|
|
"epoch": 0.5718547986076579,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9352,
|
|
"mean_token_accuracy": 0.07985315136611462,
|
|
"num_tokens": 1130046.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.154598426818847,
|
|
"epoch": 0.5768274490303331,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.913,
|
|
"mean_token_accuracy": 0.08340204060077668,
|
|
"num_tokens": 1138946.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.361656951904298,
|
|
"epoch": 0.5818000994530085,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.7485,
|
|
"mean_token_accuracy": 0.09545417353510857,
|
|
"num_tokens": 1149440.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.243297576904297,
|
|
"epoch": 0.5867727498756837,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.9164,
|
|
"mean_token_accuracy": 0.08602936565876007,
|
|
"num_tokens": 1158636.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.125365924835204,
|
|
"epoch": 0.591745400298359,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.7901,
|
|
"mean_token_accuracy": 0.09125298783183097,
|
|
"num_tokens": 1167776.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.234030151367188,
|
|
"epoch": 0.5967180507210343,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.8142,
|
|
"mean_token_accuracy": 0.08615061044692993,
|
|
"num_tokens": 1177919.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.142181634902954,
|
|
"epoch": 0.6016907011437096,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.8162,
|
|
"mean_token_accuracy": 0.09673612490296364,
|
|
"num_tokens": 1187363.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.359734725952148,
|
|
"epoch": 0.6066633515663848,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 8.0163,
|
|
"mean_token_accuracy": 0.08007314652204514,
|
|
"num_tokens": 1197329.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.210324144363403,
|
|
"epoch": 0.6116360019890602,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.9194,
|
|
"mean_token_accuracy": 0.0824542798101902,
|
|
"num_tokens": 1207408.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.167593669891357,
|
|
"epoch": 0.6166086524117355,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.9066,
|
|
"mean_token_accuracy": 0.08329624831676483,
|
|
"num_tokens": 1219258.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.238018608093261,
|
|
"epoch": 0.6215813028344107,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.8914,
|
|
"mean_token_accuracy": 0.08404905647039414,
|
|
"num_tokens": 1228561.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.111460828781128,
|
|
"epoch": 0.6265539532570861,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.8313,
|
|
"mean_token_accuracy": 0.08610942959785461,
|
|
"num_tokens": 1239665.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.152514457702637,
|
|
"epoch": 0.6315266036797613,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.8463,
|
|
"mean_token_accuracy": 0.08551793694496154,
|
|
"num_tokens": 1249490.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.17859344482422,
|
|
"epoch": 0.6364992541024366,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.8593,
|
|
"mean_token_accuracy": 0.08357597589492798,
|
|
"num_tokens": 1259328.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.283420848846436,
|
|
"epoch": 0.6414719045251119,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.8091,
|
|
"mean_token_accuracy": 0.08467675819993019,
|
|
"num_tokens": 1269523.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.04047703742981,
|
|
"epoch": 0.6464445549477872,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.8652,
|
|
"mean_token_accuracy": 0.08439442962408066,
|
|
"num_tokens": 1277961.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.285121870040893,
|
|
"epoch": 0.6514172053704624,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.8504,
|
|
"mean_token_accuracy": 0.08624732792377472,
|
|
"num_tokens": 1288228.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.085236644744873,
|
|
"epoch": 0.6563898557931377,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7828,
|
|
"mean_token_accuracy": 0.08678266704082489,
|
|
"num_tokens": 1299636.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.120037984848022,
|
|
"epoch": 0.661362506215813,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.8004,
|
|
"mean_token_accuracy": 0.08794604614377022,
|
|
"num_tokens": 1310035.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.082569122314453,
|
|
"epoch": 0.6663351566384883,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.8312,
|
|
"mean_token_accuracy": 0.09141349494457245,
|
|
"num_tokens": 1320751.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 8.134233331680297,
|
|
"epoch": 0.6713078070611636,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7128,
|
|
"mean_token_accuracy": 0.09502863362431527,
|
|
"num_tokens": 1331023.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 8.086916255950928,
|
|
"epoch": 0.6762804574838389,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.8011,
|
|
"mean_token_accuracy": 0.08197866380214691,
|
|
"num_tokens": 1341814.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 8.112795162200928,
|
|
"epoch": 0.6812531079065142,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6898,
|
|
"mean_token_accuracy": 0.09163091257214546,
|
|
"num_tokens": 1351026.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.14989185333252,
|
|
"epoch": 0.6862257583291894,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.8996,
|
|
"mean_token_accuracy": 0.07762215696275235,
|
|
"num_tokens": 1361004.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 8.034332180023194,
|
|
"epoch": 0.6911984087518648,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7494,
|
|
"mean_token_accuracy": 0.08716261237859727,
|
|
"num_tokens": 1370729.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 8.083207607269287,
|
|
"epoch": 0.69617105917454,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.7255,
|
|
"mean_token_accuracy": 0.0870910868048668,
|
|
"num_tokens": 1379525.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.957210445404053,
|
|
"epoch": 0.7011437095972153,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7618,
|
|
"mean_token_accuracy": 0.09080325812101364,
|
|
"num_tokens": 1389863.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 8.131938791275024,
|
|
"epoch": 0.7061163600198906,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.8828,
|
|
"mean_token_accuracy": 0.08380952700972558,
|
|
"num_tokens": 1399468.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.949442291259766,
|
|
"epoch": 0.7110890104425659,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6948,
|
|
"mean_token_accuracy": 0.0940646693110466,
|
|
"num_tokens": 1408785.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 8.000400161743164,
|
|
"epoch": 0.7160616608652411,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.7117,
|
|
"mean_token_accuracy": 0.09171824455261231,
|
|
"num_tokens": 1418666.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 8.186839485168457,
|
|
"epoch": 0.7210343112879165,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.8242,
|
|
"mean_token_accuracy": 0.09043069928884506,
|
|
"num_tokens": 1428699.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 8.019271898269654,
|
|
"epoch": 0.7260069617105918,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.7386,
|
|
"mean_token_accuracy": 0.08939997553825378,
|
|
"num_tokens": 1437354.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 8.042302370071411,
|
|
"epoch": 0.730979612133267,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.7591,
|
|
"mean_token_accuracy": 0.08962323293089866,
|
|
"num_tokens": 1447327.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 8.099817705154418,
|
|
"epoch": 0.7359522625559424,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.7999,
|
|
"mean_token_accuracy": 0.09006209298968315,
|
|
"num_tokens": 1457793.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.951085233688355,
|
|
"epoch": 0.7409249129786176,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.7091,
|
|
"mean_token_accuracy": 0.09049957096576691,
|
|
"num_tokens": 1467143.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 8.054324007034301,
|
|
"epoch": 0.7458975634012929,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.7436,
|
|
"mean_token_accuracy": 0.0868989959359169,
|
|
"num_tokens": 1476422.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.982741785049439,
|
|
"epoch": 0.7508702138239681,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.6828,
|
|
"mean_token_accuracy": 0.09940663874149322,
|
|
"num_tokens": 1486253.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.976361560821533,
|
|
"epoch": 0.7558428642466435,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.6932,
|
|
"mean_token_accuracy": 0.0911173690110445,
|
|
"num_tokens": 1496080.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.976806545257569,
|
|
"epoch": 0.7608155146693187,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6559,
|
|
"mean_token_accuracy": 0.09429258331656457,
|
|
"num_tokens": 1505532.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 8.003704738616943,
|
|
"epoch": 0.765788165091994,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.6102,
|
|
"mean_token_accuracy": 0.0926941230893135,
|
|
"num_tokens": 1515705.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.872346496582031,
|
|
"epoch": 0.7707608155146694,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.6424,
|
|
"mean_token_accuracy": 0.09104666113853455,
|
|
"num_tokens": 1525651.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.989510869979858,
|
|
"epoch": 0.7757334659373446,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.6179,
|
|
"mean_token_accuracy": 0.09006042331457138,
|
|
"num_tokens": 1534576.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.934913539886475,
|
|
"epoch": 0.7807061163600199,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.488,
|
|
"mean_token_accuracy": 0.10270283967256547,
|
|
"num_tokens": 1544765.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.897942733764649,
|
|
"epoch": 0.7856787667826952,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.6135,
|
|
"mean_token_accuracy": 0.09766546785831451,
|
|
"num_tokens": 1553636.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.949420356750489,
|
|
"epoch": 0.7906514172053705,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.5908,
|
|
"mean_token_accuracy": 0.09204349145293236,
|
|
"num_tokens": 1563356.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.791548585891723,
|
|
"epoch": 0.7956240676280457,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.606,
|
|
"mean_token_accuracy": 0.09886129572987556,
|
|
"num_tokens": 1572508.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.962732410430908,
|
|
"epoch": 0.8005967180507211,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.6313,
|
|
"mean_token_accuracy": 0.09463887512683869,
|
|
"num_tokens": 1581583.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.957588052749633,
|
|
"epoch": 0.8055693684733963,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.6262,
|
|
"mean_token_accuracy": 0.09475942105054855,
|
|
"num_tokens": 1592114.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.837848281860351,
|
|
"epoch": 0.8105420188960716,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.5169,
|
|
"mean_token_accuracy": 0.0997062899172306,
|
|
"num_tokens": 1602417.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.859059190750122,
|
|
"epoch": 0.8155146693187469,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.6001,
|
|
"mean_token_accuracy": 0.09124608859419822,
|
|
"num_tokens": 1611753.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.919077014923095,
|
|
"epoch": 0.8204873197414222,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.6719,
|
|
"mean_token_accuracy": 0.09552423506975175,
|
|
"num_tokens": 1621959.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.8648748874664305,
|
|
"epoch": 0.8254599701640974,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.5882,
|
|
"mean_token_accuracy": 0.0931095041334629,
|
|
"num_tokens": 1630909.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.898569536209107,
|
|
"epoch": 0.8304326205867727,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.5774,
|
|
"mean_token_accuracy": 0.09536047205328942,
|
|
"num_tokens": 1641316.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.789081382751465,
|
|
"epoch": 0.8354052710094481,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.4753,
|
|
"mean_token_accuracy": 0.10203570201992988,
|
|
"num_tokens": 1651506.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.7330341815948485,
|
|
"epoch": 0.8403779214321233,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.468,
|
|
"mean_token_accuracy": 0.09837703853845596,
|
|
"num_tokens": 1659871.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.998170804977417,
|
|
"epoch": 0.8453505718547986,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.6687,
|
|
"mean_token_accuracy": 0.08796224296092987,
|
|
"num_tokens": 1670178.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.850816154479981,
|
|
"epoch": 0.8503232222774739,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5711,
|
|
"mean_token_accuracy": 0.09666412323713303,
|
|
"num_tokens": 1680131.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.757242822647095,
|
|
"epoch": 0.8552958727001492,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.4708,
|
|
"mean_token_accuracy": 0.09963822215795518,
|
|
"num_tokens": 1689568.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.942859315872193,
|
|
"epoch": 0.8602685231228244,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5267,
|
|
"mean_token_accuracy": 0.10697205737233162,
|
|
"num_tokens": 1698039.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.757676982879639,
|
|
"epoch": 0.8652411735454998,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.5359,
|
|
"mean_token_accuracy": 0.09895949959754943,
|
|
"num_tokens": 1707659.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.809637451171875,
|
|
"epoch": 0.870213823968175,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.5247,
|
|
"mean_token_accuracy": 0.10059929713606834,
|
|
"num_tokens": 1716952.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.862903022766114,
|
|
"epoch": 0.8751864743908503,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.562,
|
|
"mean_token_accuracy": 0.09478862844407558,
|
|
"num_tokens": 1726323.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.839666557312012,
|
|
"epoch": 0.8801591248135257,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5568,
|
|
"mean_token_accuracy": 0.093773003667593,
|
|
"num_tokens": 1736265.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.794053173065185,
|
|
"epoch": 0.8851317752362009,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5303,
|
|
"mean_token_accuracy": 0.09586519375443459,
|
|
"num_tokens": 1745150.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.859850168228149,
|
|
"epoch": 0.8901044256588762,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.6115,
|
|
"mean_token_accuracy": 0.09495341330766678,
|
|
"num_tokens": 1754641.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.658648729324341,
|
|
"epoch": 0.8950770760815515,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.4228,
|
|
"mean_token_accuracy": 0.10544339269399643,
|
|
"num_tokens": 1763468.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.818254232406616,
|
|
"epoch": 0.9000497265042268,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.5498,
|
|
"mean_token_accuracy": 0.09819675534963608,
|
|
"num_tokens": 1774845.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.732444143295288,
|
|
"epoch": 0.905022376926902,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.4874,
|
|
"mean_token_accuracy": 0.09764616191387177,
|
|
"num_tokens": 1784606.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.762696981430054,
|
|
"epoch": 0.9099950273495774,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.5197,
|
|
"mean_token_accuracy": 0.09776233658194541,
|
|
"num_tokens": 1793893.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.759170389175415,
|
|
"epoch": 0.9149676777722526,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4864,
|
|
"mean_token_accuracy": 0.1026741810142994,
|
|
"num_tokens": 1804095.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.847332000732422,
|
|
"epoch": 0.9199403281949279,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.4962,
|
|
"mean_token_accuracy": 0.09903053939342499,
|
|
"num_tokens": 1813475.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.590861415863037,
|
|
"epoch": 0.9249129786176031,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.4214,
|
|
"mean_token_accuracy": 0.10564838871359825,
|
|
"num_tokens": 1822779.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.834741497039795,
|
|
"epoch": 0.9298856290402785,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4718,
|
|
"mean_token_accuracy": 0.09719581454992295,
|
|
"num_tokens": 1832239.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.63415846824646,
|
|
"epoch": 0.9348582794629537,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.4419,
|
|
"mean_token_accuracy": 0.09387974292039872,
|
|
"num_tokens": 1843395.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.717109155654907,
|
|
"epoch": 0.939830929885629,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.5299,
|
|
"mean_token_accuracy": 0.0993468426167965,
|
|
"num_tokens": 1853861.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.784553384780883,
|
|
"epoch": 0.9448035803083044,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.4638,
|
|
"mean_token_accuracy": 0.10380829498171806,
|
|
"num_tokens": 1862992.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.70339937210083,
|
|
"epoch": 0.9497762307309796,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.4693,
|
|
"mean_token_accuracy": 0.09729222431778908,
|
|
"num_tokens": 1871684.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.732805299758911,
|
|
"epoch": 0.9547488811536549,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.4873,
|
|
"mean_token_accuracy": 0.09647738561034203,
|
|
"num_tokens": 1881956.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.718251419067383,
|
|
"epoch": 0.9597215315763302,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.5568,
|
|
"mean_token_accuracy": 0.09375276640057564,
|
|
"num_tokens": 1891860.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.731233787536621,
|
|
"epoch": 0.9646941819990055,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.4563,
|
|
"mean_token_accuracy": 0.09947184100747108,
|
|
"num_tokens": 1901521.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.708991575241089,
|
|
"epoch": 0.9696668324216807,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.4339,
|
|
"mean_token_accuracy": 0.10033734291791915,
|
|
"num_tokens": 1910438.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.593982934951782,
|
|
"epoch": 0.9746394828443561,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.4344,
|
|
"mean_token_accuracy": 0.10258711129426956,
|
|
"num_tokens": 1922045.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.7459498882293705,
|
|
"epoch": 0.9796121332670313,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.2872,
|
|
"mean_token_accuracy": 0.10616242960095405,
|
|
"num_tokens": 1930330.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.534809637069702,
|
|
"epoch": 0.9845847836897066,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.2596,
|
|
"mean_token_accuracy": 0.1091254822909832,
|
|
"num_tokens": 1939485.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.7126853466033936,
|
|
"epoch": 0.989557434112382,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.3957,
|
|
"mean_token_accuracy": 0.0969047375023365,
|
|
"num_tokens": 1949448.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.716427373886108,
|
|
"epoch": 0.9945300845350572,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.4736,
|
|
"mean_token_accuracy": 0.09623041301965714,
|
|
"num_tokens": 1959923.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9945300845350572,
|
|
"eval_entropy": 7.488209557423394,
|
|
"eval_loss": 7.501986980438232,
|
|
"eval_mean_token_accuracy": 0.09682262113772779,
|
|
"eval_num_tokens": 1959923.0,
|
|
"eval_runtime": 1.9373,
|
|
"eval_samples_per_second": 1788.613,
|
|
"eval_steps_per_second": 224.028,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.548803853988647,
|
|
"epoch": 0.9995027349577325,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999997830922735,
|
|
"loss": 7.413,
|
|
"mean_token_accuracy": 0.10569150224328042,
|
|
"num_tokens": 1969105.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.66062773598565,
|
|
"epoch": 1.0039781203381402,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999989019053515,
|
|
"loss": 7.2661,
|
|
"mean_token_accuracy": 0.10590489125914043,
|
|
"num_tokens": 1979124.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.671492671966552,
|
|
"epoch": 1.0089507707608154,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999973428851536,
|
|
"loss": 7.2748,
|
|
"mean_token_accuracy": 0.10845026299357415,
|
|
"num_tokens": 1987813.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.5896931171417235,
|
|
"epoch": 1.0139234211834909,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999951060363766,
|
|
"loss": 7.2664,
|
|
"mean_token_accuracy": 0.10322905778884887,
|
|
"num_tokens": 1996743.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.6094602108001705,
|
|
"epoch": 1.0188960716061661,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999921913657592,
|
|
"loss": 7.2511,
|
|
"mean_token_accuracy": 0.10825500115752221,
|
|
"num_tokens": 2006499.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.573673152923584,
|
|
"epoch": 1.0238687220288414,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999885988820821,
|
|
"loss": 7.2413,
|
|
"mean_token_accuracy": 0.10702010616660118,
|
|
"num_tokens": 2015103.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.60667405128479,
|
|
"epoch": 1.0288413724515166,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999843285961683,
|
|
"loss": 7.1908,
|
|
"mean_token_accuracy": 0.10732990354299546,
|
|
"num_tokens": 2024681.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.556708955764771,
|
|
"epoch": 1.0338140228741919,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999793805208822,
|
|
"loss": 7.2574,
|
|
"mean_token_accuracy": 0.10310408398509026,
|
|
"num_tokens": 2034388.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.498585605621338,
|
|
"epoch": 1.0387866732968671,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999737546711305,
|
|
"loss": 7.189,
|
|
"mean_token_accuracy": 0.10633601993322372,
|
|
"num_tokens": 2042729.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.61358380317688,
|
|
"epoch": 1.0437593237195426,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999674510638618,
|
|
"loss": 7.2889,
|
|
"mean_token_accuracy": 0.10114088803529739,
|
|
"num_tokens": 2051669.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.553178501129151,
|
|
"epoch": 1.0487319741422179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999604697180661,
|
|
"loss": 7.2269,
|
|
"mean_token_accuracy": 0.11130643114447594,
|
|
"num_tokens": 2061002.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.503707408905029,
|
|
"epoch": 1.053704624564893,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999528106547759,
|
|
"loss": 7.1384,
|
|
"mean_token_accuracy": 0.11123101264238358,
|
|
"num_tokens": 2070222.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.556910037994385,
|
|
"epoch": 1.0586772749875684,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999444738970644,
|
|
"loss": 7.329,
|
|
"mean_token_accuracy": 0.10767423883080482,
|
|
"num_tokens": 2079683.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.55855360031128,
|
|
"epoch": 1.0636499254102436,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999354594700474,
|
|
"loss": 7.2931,
|
|
"mean_token_accuracy": 0.10593407750129699,
|
|
"num_tokens": 2089659.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.590776634216309,
|
|
"epoch": 1.0686225758329189,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999257674008817,
|
|
"loss": 7.1852,
|
|
"mean_token_accuracy": 0.11362824738025665,
|
|
"num_tokens": 2099766.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.531158494949341,
|
|
"epoch": 1.0735952262555943,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999153977187656,
|
|
"loss": 7.3396,
|
|
"mean_token_accuracy": 0.09754043966531753,
|
|
"num_tokens": 2110244.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.652919435501099,
|
|
"epoch": 1.0785678766782696,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000499904350454939,
|
|
"loss": 7.2637,
|
|
"mean_token_accuracy": 0.10820004642009735,
|
|
"num_tokens": 2120068.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.437949848175049,
|
|
"epoch": 1.0835405271009448,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998926256426829,
|
|
"loss": 7.2406,
|
|
"mean_token_accuracy": 0.10829075872898102,
|
|
"num_tokens": 2128862.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.683877944946289,
|
|
"epoch": 1.08851317752362,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998802233173196,
|
|
"loss": 7.3452,
|
|
"mean_token_accuracy": 0.09526378959417343,
|
|
"num_tokens": 2140618.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.499522924423218,
|
|
"epoch": 1.0934858279462953,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998671435162123,
|
|
"loss": 7.2207,
|
|
"mean_token_accuracy": 0.10381940528750419,
|
|
"num_tokens": 2151890.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.5036125659942625,
|
|
"epoch": 1.0984584783689706,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998533862787657,
|
|
"loss": 7.1895,
|
|
"mean_token_accuracy": 0.11296708136796951,
|
|
"num_tokens": 2160862.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.519911384582519,
|
|
"epoch": 1.1034311287916458,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998389516464244,
|
|
"loss": 7.2329,
|
|
"mean_token_accuracy": 0.11361916735768318,
|
|
"num_tokens": 2171631.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.524933004379273,
|
|
"epoch": 1.1084037792143213,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998238396626746,
|
|
"loss": 7.1747,
|
|
"mean_token_accuracy": 0.10848848670721054,
|
|
"num_tokens": 2181564.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.52897162437439,
|
|
"epoch": 1.1133764296369966,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998080503730427,
|
|
"loss": 7.2283,
|
|
"mean_token_accuracy": 0.11038302779197692,
|
|
"num_tokens": 2191288.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.483823251724243,
|
|
"epoch": 1.1183490800596718,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997915838250956,
|
|
"loss": 7.183,
|
|
"mean_token_accuracy": 0.10745534673333168,
|
|
"num_tokens": 2201254.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.4095738410949705,
|
|
"epoch": 1.123321730482347,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997744400684406,
|
|
"loss": 7.1587,
|
|
"mean_token_accuracy": 0.11007041931152343,
|
|
"num_tokens": 2210325.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.51028847694397,
|
|
"epoch": 1.1282943809050223,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997566191547251,
|
|
"loss": 7.237,
|
|
"mean_token_accuracy": 0.10701308622956276,
|
|
"num_tokens": 2220357.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.440891551971435,
|
|
"epoch": 1.1332670313276976,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997381211376364,
|
|
"loss": 7.1617,
|
|
"mean_token_accuracy": 0.11563558354973794,
|
|
"num_tokens": 2229208.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.447800922393799,
|
|
"epoch": 1.138239681750373,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004997189460729019,
|
|
"loss": 7.1472,
|
|
"mean_token_accuracy": 0.11236186400055885,
|
|
"num_tokens": 2238780.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.392899417877198,
|
|
"epoch": 1.1432123321730483,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004996990940182884,
|
|
"loss": 7.1841,
|
|
"mean_token_accuracy": 0.11078342124819755,
|
|
"num_tokens": 2249457.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.465713930130005,
|
|
"epoch": 1.1481849825957235,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004996785650336024,
|
|
"loss": 7.1892,
|
|
"mean_token_accuracy": 0.10895669460296631,
|
|
"num_tokens": 2259565.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.375476360321045,
|
|
"epoch": 1.1531576330183988,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996573591806898,
|
|
"loss": 7.0812,
|
|
"mean_token_accuracy": 0.11717669442296028,
|
|
"num_tokens": 2268301.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.478143262863159,
|
|
"epoch": 1.158130283441074,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996354765234353,
|
|
"loss": 7.2115,
|
|
"mean_token_accuracy": 0.11243066862225533,
|
|
"num_tokens": 2278298.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.434989595413208,
|
|
"epoch": 1.1631029338637493,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004996129171277628,
|
|
"loss": 7.1646,
|
|
"mean_token_accuracy": 0.11214689686894416,
|
|
"num_tokens": 2288442.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.3911176204681395,
|
|
"epoch": 1.1680755842864246,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995896810616349,
|
|
"loss": 7.0728,
|
|
"mean_token_accuracy": 0.1135096326470375,
|
|
"num_tokens": 2298785.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.457903289794922,
|
|
"epoch": 1.1730482347091,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004995657683950528,
|
|
"loss": 7.2192,
|
|
"mean_token_accuracy": 0.10406702384352684,
|
|
"num_tokens": 2308881.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.355658769607544,
|
|
"epoch": 1.1780208851317753,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995411792000559,
|
|
"loss": 7.119,
|
|
"mean_token_accuracy": 0.1158958025276661,
|
|
"num_tokens": 2318447.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.451979303359986,
|
|
"epoch": 1.1829935355544505,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995159135507218,
|
|
"loss": 7.119,
|
|
"mean_token_accuracy": 0.12004166394472122,
|
|
"num_tokens": 2327162.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.467351293563842,
|
|
"epoch": 1.1879661859771258,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499489971523166,
|
|
"loss": 7.167,
|
|
"mean_token_accuracy": 0.10660808980464935,
|
|
"num_tokens": 2337207.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.33328948020935,
|
|
"epoch": 1.192938836399801,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004994633531955415,
|
|
"loss": 7.1606,
|
|
"mean_token_accuracy": 0.10822098404169082,
|
|
"num_tokens": 2347381.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.472771406173706,
|
|
"epoch": 1.1979114868224765,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004994360586480391,
|
|
"loss": 7.1881,
|
|
"mean_token_accuracy": 0.11179639548063278,
|
|
"num_tokens": 2357942.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.356680679321289,
|
|
"epoch": 1.2028841372451518,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994080879628864,
|
|
"loss": 7.1036,
|
|
"mean_token_accuracy": 0.11353320553898812,
|
|
"num_tokens": 2367242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.503470373153687,
|
|
"epoch": 1.207856787667827,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004993794412243481,
|
|
"loss": 7.1731,
|
|
"mean_token_accuracy": 0.11843622848391533,
|
|
"num_tokens": 2376344.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.321029806137085,
|
|
"epoch": 1.2128294380905023,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004993501185187255,
|
|
"loss": 7.175,
|
|
"mean_token_accuracy": 0.11539561301469803,
|
|
"num_tokens": 2387206.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.5107738971710205,
|
|
"epoch": 1.2178020885131775,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004993201199343565,
|
|
"loss": 7.2087,
|
|
"mean_token_accuracy": 0.11212565749883652,
|
|
"num_tokens": 2397532.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.33886947631836,
|
|
"epoch": 1.2227747389358528,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004992894455616149,
|
|
"loss": 7.0622,
|
|
"mean_token_accuracy": 0.10937203243374824,
|
|
"num_tokens": 2409165.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.425497341156006,
|
|
"epoch": 1.227747389358528,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004992580954929108,
|
|
"loss": 7.1683,
|
|
"mean_token_accuracy": 0.11230182573199272,
|
|
"num_tokens": 2420657.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.326796674728394,
|
|
"epoch": 1.2327200397812033,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004992260698226894,
|
|
"loss": 7.1024,
|
|
"mean_token_accuracy": 0.11296775862574578,
|
|
"num_tokens": 2429936.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.369725608825684,
|
|
"epoch": 1.2376926902038787,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991933686474315,
|
|
"loss": 7.0853,
|
|
"mean_token_accuracy": 0.11595127657055855,
|
|
"num_tokens": 2440386.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.347142744064331,
|
|
"epoch": 1.242665340626554,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004991599920656529,
|
|
"loss": 7.2428,
|
|
"mean_token_accuracy": 0.11036591380834579,
|
|
"num_tokens": 2449086.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.615409803390503,
|
|
"epoch": 1.2476379910492292,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991259401779042,
|
|
"loss": 7.2543,
|
|
"mean_token_accuracy": 0.10894604101777076,
|
|
"num_tokens": 2459380.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.305810117721558,
|
|
"epoch": 1.2526106414719045,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004990912130867704,
|
|
"loss": 7.0957,
|
|
"mean_token_accuracy": 0.1171362578868866,
|
|
"num_tokens": 2468623.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.462665271759033,
|
|
"epoch": 1.2575832918945797,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004990558108968703,
|
|
"loss": 7.0733,
|
|
"mean_token_accuracy": 0.11163564622402192,
|
|
"num_tokens": 2479817.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.359383773803711,
|
|
"epoch": 1.2625559423172552,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004990197337148571,
|
|
"loss": 7.086,
|
|
"mean_token_accuracy": 0.11412426605820655,
|
|
"num_tokens": 2490544.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.300520944595337,
|
|
"epoch": 1.2675285927399305,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000498982981649417,
|
|
"loss": 7.0359,
|
|
"mean_token_accuracy": 0.12793442085385323,
|
|
"num_tokens": 2500516.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.379915714263916,
|
|
"epoch": 1.2725012431626057,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004989455548112695,
|
|
"loss": 7.177,
|
|
"mean_token_accuracy": 0.11028578877449036,
|
|
"num_tokens": 2510508.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.4297041416168215,
|
|
"epoch": 1.277473893585281,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989074533131671,
|
|
"loss": 7.1298,
|
|
"mean_token_accuracy": 0.11257019862532616,
|
|
"num_tokens": 2520532.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.364310598373413,
|
|
"epoch": 1.2824465440079562,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004988686772698945,
|
|
"loss": 7.0823,
|
|
"mean_token_accuracy": 0.11520142257213592,
|
|
"num_tokens": 2529912.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.360514116287232,
|
|
"epoch": 1.2874191944306315,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004988292267982689,
|
|
"loss": 7.1672,
|
|
"mean_token_accuracy": 0.11030791848897933,
|
|
"num_tokens": 2540700.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.358193731307983,
|
|
"epoch": 1.2923918448533067,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004987891020171387,
|
|
"loss": 7.0408,
|
|
"mean_token_accuracy": 0.1187594324350357,
|
|
"num_tokens": 2549343.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.262971496582031,
|
|
"epoch": 1.297364495275982,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004987483030473845,
|
|
"loss": 7.0898,
|
|
"mean_token_accuracy": 0.11714393496513367,
|
|
"num_tokens": 2558630.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.371310091018676,
|
|
"epoch": 1.3023371456986574,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987068300119173,
|
|
"loss": 7.078,
|
|
"mean_token_accuracy": 0.11369528621435165,
|
|
"num_tokens": 2568559.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.3390827655792235,
|
|
"epoch": 1.3073097961213327,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004986646830356792,
|
|
"loss": 7.0962,
|
|
"mean_token_accuracy": 0.11881779953837394,
|
|
"num_tokens": 2578088.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.237670850753784,
|
|
"epoch": 1.312282446544008,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004986218622456423,
|
|
"loss": 7.0736,
|
|
"mean_token_accuracy": 0.11693579256534577,
|
|
"num_tokens": 2588518.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.485841464996338,
|
|
"epoch": 1.3172550969666832,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000498578367770809,
|
|
"loss": 7.1883,
|
|
"mean_token_accuracy": 0.10811706483364106,
|
|
"num_tokens": 2598850.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.35091381072998,
|
|
"epoch": 1.3222277473893587,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004985341997422111,
|
|
"loss": 7.183,
|
|
"mean_token_accuracy": 0.10256853848695754,
|
|
"num_tokens": 2610667.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.325898838043213,
|
|
"epoch": 1.327200397812034,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004984893582929093,
|
|
"loss": 7.0341,
|
|
"mean_token_accuracy": 0.12277554646134377,
|
|
"num_tokens": 2620430.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.2079283714294435,
|
|
"epoch": 1.3321730482347092,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004984438435579933,
|
|
"loss": 6.9044,
|
|
"mean_token_accuracy": 0.1257750518620014,
|
|
"num_tokens": 2629433.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.216365194320678,
|
|
"epoch": 1.3371456986573844,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004983976556745812,
|
|
"loss": 7.1079,
|
|
"mean_token_accuracy": 0.10878997221589089,
|
|
"num_tokens": 2639611.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.397309875488281,
|
|
"epoch": 1.3421183490800597,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000498350794781819,
|
|
"loss": 7.1016,
|
|
"mean_token_accuracy": 0.11138525605201721,
|
|
"num_tokens": 2649494.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.319606161117553,
|
|
"epoch": 1.347090999502735,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004983032610208801,
|
|
"loss": 7.011,
|
|
"mean_token_accuracy": 0.11263529062271119,
|
|
"num_tokens": 2658723.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.232192325592041,
|
|
"epoch": 1.3520636499254102,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004982550545349649,
|
|
"loss": 7.0587,
|
|
"mean_token_accuracy": 0.1151237927377224,
|
|
"num_tokens": 2668530.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.411221837997436,
|
|
"epoch": 1.3570363003480854,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004982061754693008,
|
|
"loss": 7.1231,
|
|
"mean_token_accuracy": 0.11438949555158615,
|
|
"num_tokens": 2678788.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.242826366424561,
|
|
"epoch": 1.362008950770761,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004981566239711412,
|
|
"loss": 7.0343,
|
|
"mean_token_accuracy": 0.11359024792909622,
|
|
"num_tokens": 2688308.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.141946029663086,
|
|
"epoch": 1.3669816011934361,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0004981064001897651,
|
|
"loss": 6.923,
|
|
"mean_token_accuracy": 0.12715769335627555,
|
|
"num_tokens": 2698146.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.367173194885254,
|
|
"epoch": 1.3719542516161114,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004980555042764772,
|
|
"loss": 7.0322,
|
|
"mean_token_accuracy": 0.11541925519704818,
|
|
"num_tokens": 2707336.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.302237701416016,
|
|
"epoch": 1.3769269020387866,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004980039363846069,
|
|
"loss": 7.0114,
|
|
"mean_token_accuracy": 0.12369872853159905,
|
|
"num_tokens": 2717573.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.241599416732788,
|
|
"epoch": 1.381899552461462,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000497951696669508,
|
|
"loss": 6.9995,
|
|
"mean_token_accuracy": 0.12224290147423744,
|
|
"num_tokens": 2727488.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.242745351791382,
|
|
"epoch": 1.3868722028841374,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004978987852885579,
|
|
"loss": 6.9943,
|
|
"mean_token_accuracy": 0.12633508816361427,
|
|
"num_tokens": 2738010.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.241048336029053,
|
|
"epoch": 1.3918448533068126,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978452024011582,
|
|
"loss": 7.0408,
|
|
"mean_token_accuracy": 0.11665860190987587,
|
|
"num_tokens": 2746984.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.3179529190063475,
|
|
"epoch": 1.3968175037294879,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000497790948168733,
|
|
"loss": 7.0649,
|
|
"mean_token_accuracy": 0.11003333255648613,
|
|
"num_tokens": 2756578.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.231767416000366,
|
|
"epoch": 1.4017901541521631,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004977360227547288,
|
|
"loss": 7.0013,
|
|
"mean_token_accuracy": 0.11939035207033158,
|
|
"num_tokens": 2765773.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.205044269561768,
|
|
"epoch": 1.4067628045748384,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004976804263246146,
|
|
"loss": 6.9899,
|
|
"mean_token_accuracy": 0.11811095997691154,
|
|
"num_tokens": 2775002.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.309245014190674,
|
|
"epoch": 1.4117354549975136,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004976241590458803,
|
|
"loss": 7.0656,
|
|
"mean_token_accuracy": 0.12084579244256019,
|
|
"num_tokens": 2784682.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.197923421859741,
|
|
"epoch": 1.4167081054201889,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004975672210880372,
|
|
"loss": 6.9604,
|
|
"mean_token_accuracy": 0.12463531345129013,
|
|
"num_tokens": 2794371.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.3131937980651855,
|
|
"epoch": 1.4216807558428641,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004975096126226171,
|
|
"loss": 7.0379,
|
|
"mean_token_accuracy": 0.11712399497628212,
|
|
"num_tokens": 2803840.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.219164562225342,
|
|
"epoch": 1.4266534062655396,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004974513338231715,
|
|
"loss": 6.9517,
|
|
"mean_token_accuracy": 0.12495556622743606,
|
|
"num_tokens": 2813536.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.154695463180542,
|
|
"epoch": 1.4316260566882149,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004973923848652715,
|
|
"loss": 7.0029,
|
|
"mean_token_accuracy": 0.11516394093632698,
|
|
"num_tokens": 2822741.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.313390064239502,
|
|
"epoch": 1.43659870711089,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004973327659265073,
|
|
"loss": 6.9737,
|
|
"mean_token_accuracy": 0.1199723333120346,
|
|
"num_tokens": 2832208.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.077762508392334,
|
|
"epoch": 1.4415713575335654,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004972724771864874,
|
|
"loss": 6.8388,
|
|
"mean_token_accuracy": 0.12911385893821717,
|
|
"num_tokens": 2841672.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.217231750488281,
|
|
"epoch": 1.4465440079562406,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004972115188268378,
|
|
"loss": 6.9823,
|
|
"mean_token_accuracy": 0.1208034448325634,
|
|
"num_tokens": 2852323.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.230539703369141,
|
|
"epoch": 1.451516658378916,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004971498910312024,
|
|
"loss": 7.0263,
|
|
"mean_token_accuracy": 0.11673628762364388,
|
|
"num_tokens": 2861863.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.258527660369873,
|
|
"epoch": 1.4564893088015913,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004970875939852412,
|
|
"loss": 6.9738,
|
|
"mean_token_accuracy": 0.1178445041179657,
|
|
"num_tokens": 2871608.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.212831115722656,
|
|
"epoch": 1.4614619592242666,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000497024627876631,
|
|
"loss": 6.9147,
|
|
"mean_token_accuracy": 0.12569650560617446,
|
|
"num_tokens": 2882091.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.180489492416382,
|
|
"epoch": 1.4664346096469418,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004969609928950639,
|
|
"loss": 6.8964,
|
|
"mean_token_accuracy": 0.12846860513091088,
|
|
"num_tokens": 2892499.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.226547574996948,
|
|
"epoch": 1.471407260069617,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004968966892322468,
|
|
"loss": 6.9416,
|
|
"mean_token_accuracy": 0.1151064857840538,
|
|
"num_tokens": 2902208.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.236787509918213,
|
|
"epoch": 1.4763799104922923,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004968317170819018,
|
|
"loss": 6.9461,
|
|
"mean_token_accuracy": 0.11949458941817284,
|
|
"num_tokens": 2911196.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.088937759399414,
|
|
"epoch": 1.4813525609149676,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004967660766397642,
|
|
"loss": 6.963,
|
|
"mean_token_accuracy": 0.11785544157028198,
|
|
"num_tokens": 2921339.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.247524166107178,
|
|
"epoch": 1.4863252113376428,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004966997681035829,
|
|
"loss": 6.925,
|
|
"mean_token_accuracy": 0.11916105672717095,
|
|
"num_tokens": 2930795.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.125017929077148,
|
|
"epoch": 1.4912978617603183,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004966327916731195,
|
|
"loss": 6.9005,
|
|
"mean_token_accuracy": 0.12182693481445313,
|
|
"num_tokens": 2939647.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.4912978617603183,
|
|
"eval_entropy": 7.001291468396165,
|
|
"eval_loss": 7.145375728607178,
|
|
"eval_mean_token_accuracy": 0.11232445977868573,
|
|
"eval_num_tokens": 2939647.0,
|
|
"eval_runtime": 1.9445,
|
|
"eval_samples_per_second": 1781.927,
|
|
"eval_steps_per_second": 223.191,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.182607412338257,
|
|
"epoch": 1.4962705121829936,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004965651475501476,
|
|
"loss": 6.9239,
|
|
"mean_token_accuracy": 0.12209872081875801,
|
|
"num_tokens": 2948365.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.105114412307739,
|
|
"epoch": 1.5012431626056688,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004964968359384524,
|
|
"loss": 6.9574,
|
|
"mean_token_accuracy": 0.11654346659779549,
|
|
"num_tokens": 2957447.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.24699444770813,
|
|
"epoch": 1.506215813028344,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004964278570438299,
|
|
"loss": 6.9374,
|
|
"mean_token_accuracy": 0.11978609785437584,
|
|
"num_tokens": 2965938.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.204099178314209,
|
|
"epoch": 1.5111884634510195,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004963582110740865,
|
|
"loss": 6.9311,
|
|
"mean_token_accuracy": 0.12392050102353096,
|
|
"num_tokens": 2975830.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.181468343734741,
|
|
"epoch": 1.5161611138736948,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004962878982390381,
|
|
"loss": 6.9359,
|
|
"mean_token_accuracy": 0.1233926922082901,
|
|
"num_tokens": 2985391.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.240592908859253,
|
|
"epoch": 1.52113376429637,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004962169187505097,
|
|
"loss": 6.9377,
|
|
"mean_token_accuracy": 0.13239589110016822,
|
|
"num_tokens": 2994397.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.134807014465332,
|
|
"epoch": 1.5261064147190453,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004961452728223343,
|
|
"loss": 6.9666,
|
|
"mean_token_accuracy": 0.11780434027314186,
|
|
"num_tokens": 3005085.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.188982009887695,
|
|
"epoch": 1.5310790651417205,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004960729606703535,
|
|
"loss": 6.9599,
|
|
"mean_token_accuracy": 0.11750580072402954,
|
|
"num_tokens": 3015500.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.229723215103149,
|
|
"epoch": 1.5360517155643958,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004959999825124147,
|
|
"loss": 7.0544,
|
|
"mean_token_accuracy": 0.11835427656769752,
|
|
"num_tokens": 3024862.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.2740918636322025,
|
|
"epoch": 1.541024365987071,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000495926338568373,
|
|
"loss": 6.9134,
|
|
"mean_token_accuracy": 0.12224830985069275,
|
|
"num_tokens": 3034775.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.102099514007568,
|
|
"epoch": 1.5459970164097463,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004958520290600884,
|
|
"loss": 6.948,
|
|
"mean_token_accuracy": 0.12044252455234528,
|
|
"num_tokens": 3044048.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.186917638778686,
|
|
"epoch": 1.5509696668324215,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004957770542114262,
|
|
"loss": 6.9563,
|
|
"mean_token_accuracy": 0.11956393122673034,
|
|
"num_tokens": 3054237.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.061102628707886,
|
|
"epoch": 1.555942317255097,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004957014142482563,
|
|
"loss": 6.7598,
|
|
"mean_token_accuracy": 0.13404591977596284,
|
|
"num_tokens": 3064787.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.064999341964722,
|
|
"epoch": 1.5609149676777723,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004956251093984521,
|
|
"loss": 6.9456,
|
|
"mean_token_accuracy": 0.11977337896823884,
|
|
"num_tokens": 3074002.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.027366781234742,
|
|
"epoch": 1.5658876181004475,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004955481398918902,
|
|
"loss": 6.8016,
|
|
"mean_token_accuracy": 0.12725746557116507,
|
|
"num_tokens": 3084270.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.232355976104737,
|
|
"epoch": 1.5708602685231228,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004954705059604495,
|
|
"loss": 6.9126,
|
|
"mean_token_accuracy": 0.12521509304642678,
|
|
"num_tokens": 3095236.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.096924591064453,
|
|
"epoch": 1.5758329189457982,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004953922078380104,
|
|
"loss": 6.9163,
|
|
"mean_token_accuracy": 0.12283464893698692,
|
|
"num_tokens": 3104912.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.113906145095825,
|
|
"epoch": 1.5808055693684735,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004953132457604544,
|
|
"loss": 6.9309,
|
|
"mean_token_accuracy": 0.12406986579298973,
|
|
"num_tokens": 3114782.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.318424797058105,
|
|
"epoch": 1.5857782197911487,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004952336199656632,
|
|
"loss": 6.9245,
|
|
"mean_token_accuracy": 0.1289089299738407,
|
|
"num_tokens": 3123606.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.943080091476441,
|
|
"epoch": 1.590750870213824,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004951533306935181,
|
|
"loss": 6.7715,
|
|
"mean_token_accuracy": 0.13360216245055198,
|
|
"num_tokens": 3133343.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.239938402175904,
|
|
"epoch": 1.5957235206364992,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004950723781858991,
|
|
"loss": 7.0037,
|
|
"mean_token_accuracy": 0.11350896582007408,
|
|
"num_tokens": 3143137.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.120584869384766,
|
|
"epoch": 1.6006961710591745,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004949907626866844,
|
|
"loss": 6.866,
|
|
"mean_token_accuracy": 0.12739269509911538,
|
|
"num_tokens": 3152595.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.101911211013794,
|
|
"epoch": 1.6056688214818498,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004949084844417493,
|
|
"loss": 6.9308,
|
|
"mean_token_accuracy": 0.12391954734921455,
|
|
"num_tokens": 3162333.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.076293897628784,
|
|
"epoch": 1.610641471904525,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004948255436989657,
|
|
"loss": 6.8529,
|
|
"mean_token_accuracy": 0.12645021006464957,
|
|
"num_tokens": 3172765.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.179982900619507,
|
|
"epoch": 1.6156141223272003,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004947419407082017,
|
|
"loss": 6.9976,
|
|
"mean_token_accuracy": 0.1176691435277462,
|
|
"num_tokens": 3183613.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.222647380828858,
|
|
"epoch": 1.6205867727498757,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004946576757213202,
|
|
"loss": 6.8875,
|
|
"mean_token_accuracy": 0.12390058711171151,
|
|
"num_tokens": 3193020.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.116787004470825,
|
|
"epoch": 1.625559423172551,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004945727489921784,
|
|
"loss": 6.8888,
|
|
"mean_token_accuracy": 0.11933976411819458,
|
|
"num_tokens": 3202845.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.103117275238037,
|
|
"epoch": 1.6305320735952262,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004944871607766273,
|
|
"loss": 6.8284,
|
|
"mean_token_accuracy": 0.1184540294110775,
|
|
"num_tokens": 3211195.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.100602722167968,
|
|
"epoch": 1.6355047240179017,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004944009113325104,
|
|
"loss": 6.9059,
|
|
"mean_token_accuracy": 0.12281278893351555,
|
|
"num_tokens": 3221549.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.072221374511718,
|
|
"epoch": 1.640477374440577,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004943140009196635,
|
|
"loss": 6.7788,
|
|
"mean_token_accuracy": 0.13065044283866883,
|
|
"num_tokens": 3232206.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.175943613052368,
|
|
"epoch": 1.6454500248632522,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004942264297999135,
|
|
"loss": 6.8858,
|
|
"mean_token_accuracy": 0.11973264366388321,
|
|
"num_tokens": 3241613.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.959410762786865,
|
|
"epoch": 1.6504226752859275,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004941381982370778,
|
|
"loss": 6.8304,
|
|
"mean_token_accuracy": 0.12584607526659966,
|
|
"num_tokens": 3251968.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.116484069824219,
|
|
"epoch": 1.6553953257086027,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004940493064969633,
|
|
"loss": 6.8165,
|
|
"mean_token_accuracy": 0.12527019307017326,
|
|
"num_tokens": 3261969.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.102724695205689,
|
|
"epoch": 1.660367976131278,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000493959754847366,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.1251097373664379,
|
|
"num_tokens": 3270894.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.12782998085022,
|
|
"epoch": 1.6653406265539532,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004938695435580698,
|
|
"loss": 6.9141,
|
|
"mean_token_accuracy": 0.11672893092036248,
|
|
"num_tokens": 3280949.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.176233005523682,
|
|
"epoch": 1.6703132769766285,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004937786729008459,
|
|
"loss": 6.8742,
|
|
"mean_token_accuracy": 0.1206704966723919,
|
|
"num_tokens": 3291449.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.020315790176392,
|
|
"epoch": 1.6752859273993037,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004936871431494517,
|
|
"loss": 6.8755,
|
|
"mean_token_accuracy": 0.12558896467089653,
|
|
"num_tokens": 3301632.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.131391334533691,
|
|
"epoch": 1.680258577821979,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004935949545796307,
|
|
"loss": 6.7433,
|
|
"mean_token_accuracy": 0.1317845582962036,
|
|
"num_tokens": 3310802.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.0085371971130375,
|
|
"epoch": 1.6852312282446544,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004935021074691106,
|
|
"loss": 6.8027,
|
|
"mean_token_accuracy": 0.12375484257936478,
|
|
"num_tokens": 3320833.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.030786848068237,
|
|
"epoch": 1.6902038786673297,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004934086020976032,
|
|
"loss": 6.7887,
|
|
"mean_token_accuracy": 0.12935223877429963,
|
|
"num_tokens": 3330030.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.9117405891418455,
|
|
"epoch": 1.695176529090005,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004933144387468037,
|
|
"loss": 6.5805,
|
|
"mean_token_accuracy": 0.1365821048617363,
|
|
"num_tokens": 3338745.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.116425371170044,
|
|
"epoch": 1.7001491795126804,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004932196177003891,
|
|
"loss": 6.8367,
|
|
"mean_token_accuracy": 0.1287825010716915,
|
|
"num_tokens": 3348067.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.968819570541382,
|
|
"epoch": 1.7051218299353557,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000493124139244018,
|
|
"loss": 6.715,
|
|
"mean_token_accuracy": 0.13502221032977105,
|
|
"num_tokens": 3357061.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.092081594467163,
|
|
"epoch": 1.710094480358031,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004930280036653294,
|
|
"loss": 6.7723,
|
|
"mean_token_accuracy": 0.12447517067193985,
|
|
"num_tokens": 3366573.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.042446565628052,
|
|
"epoch": 1.7150671307807062,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000492931211253942,
|
|
"loss": 6.8374,
|
|
"mean_token_accuracy": 0.120162034034729,
|
|
"num_tokens": 3376996.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.199119901657104,
|
|
"epoch": 1.7200397812033814,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004928337623014535,
|
|
"loss": 6.8617,
|
|
"mean_token_accuracy": 0.12127063050866127,
|
|
"num_tokens": 3387615.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.900438976287842,
|
|
"epoch": 1.7250124316260567,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000492735657101439,
|
|
"loss": 6.645,
|
|
"mean_token_accuracy": 0.1389951892197132,
|
|
"num_tokens": 3397578.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.961544942855835,
|
|
"epoch": 1.729985082048732,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004926368959494513,
|
|
"loss": 6.826,
|
|
"mean_token_accuracy": 0.12280063703656197,
|
|
"num_tokens": 3406697.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.196460247039795,
|
|
"epoch": 1.7349577324714072,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004925374791430186,
|
|
"loss": 6.7961,
|
|
"mean_token_accuracy": 0.13026892021298409,
|
|
"num_tokens": 3416631.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.0557269096374515,
|
|
"epoch": 1.7399303828940824,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004924374069816448,
|
|
"loss": 6.8855,
|
|
"mean_token_accuracy": 0.1204845979809761,
|
|
"num_tokens": 3426564.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.084575223922729,
|
|
"epoch": 1.7449030333167577,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004923366797668081,
|
|
"loss": 6.8742,
|
|
"mean_token_accuracy": 0.1259743705391884,
|
|
"num_tokens": 3436200.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.0653331756591795,
|
|
"epoch": 1.7498756837394331,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004922352978019597,
|
|
"loss": 6.8491,
|
|
"mean_token_accuracy": 0.12475225552916527,
|
|
"num_tokens": 3446460.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.123024749755859,
|
|
"epoch": 1.7548483341621084,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0004921332613925239,
|
|
"loss": 6.7246,
|
|
"mean_token_accuracy": 0.13125914633274077,
|
|
"num_tokens": 3457629.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.050250101089477,
|
|
"epoch": 1.7598209845847836,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004920305708458964,
|
|
"loss": 6.8768,
|
|
"mean_token_accuracy": 0.12656931802630425,
|
|
"num_tokens": 3467754.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.017735004425049,
|
|
"epoch": 1.7647936350074591,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004919272264714431,
|
|
"loss": 6.8203,
|
|
"mean_token_accuracy": 0.12794173061847686,
|
|
"num_tokens": 3477094.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.077376747131348,
|
|
"epoch": 1.7697662854301344,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004918232285805004,
|
|
"loss": 6.775,
|
|
"mean_token_accuracy": 0.13335634395480156,
|
|
"num_tokens": 3487276.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.10619854927063,
|
|
"epoch": 1.7747389358528096,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004917185774863727,
|
|
"loss": 6.799,
|
|
"mean_token_accuracy": 0.12840666621923447,
|
|
"num_tokens": 3498118.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.950999164581299,
|
|
"epoch": 1.7797115862754849,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004916132735043329,
|
|
"loss": 6.6056,
|
|
"mean_token_accuracy": 0.1325760453939438,
|
|
"num_tokens": 3508286.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.952149963378906,
|
|
"epoch": 1.7846842366981601,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004915073169516204,
|
|
"loss": 6.7335,
|
|
"mean_token_accuracy": 0.13554469794034957,
|
|
"num_tokens": 3517979.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.96529369354248,
|
|
"epoch": 1.7896568871208354,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004914007081474406,
|
|
"loss": 6.7736,
|
|
"mean_token_accuracy": 0.13011404648423194,
|
|
"num_tokens": 3528003.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.123687791824341,
|
|
"epoch": 1.7946295375435106,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004912934474129641,
|
|
"loss": 6.8397,
|
|
"mean_token_accuracy": 0.12725987210869788,
|
|
"num_tokens": 3537932.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.9904521942138675,
|
|
"epoch": 1.7996021879661859,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004911855350713251,
|
|
"loss": 6.9592,
|
|
"mean_token_accuracy": 0.12082737982273102,
|
|
"num_tokens": 3548584.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.103420209884644,
|
|
"epoch": 1.8045748383888611,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004910769714476214,
|
|
"loss": 6.7673,
|
|
"mean_token_accuracy": 0.13383484184741973,
|
|
"num_tokens": 3558117.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.989352369308472,
|
|
"epoch": 1.8095474888115366,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004909677568689122,
|
|
"loss": 6.8914,
|
|
"mean_token_accuracy": 0.11944287568330765,
|
|
"num_tokens": 3569226.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.100991725921631,
|
|
"epoch": 1.8145201392342118,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004908578916642183,
|
|
"loss": 6.7524,
|
|
"mean_token_accuracy": 0.12712686210870744,
|
|
"num_tokens": 3578375.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.031591796875,
|
|
"epoch": 1.819492789656887,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004907473761645205,
|
|
"loss": 6.761,
|
|
"mean_token_accuracy": 0.1328805297613144,
|
|
"num_tokens": 3587356.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.0302153587341305,
|
|
"epoch": 1.8244654400795626,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004906362107027583,
|
|
"loss": 6.852,
|
|
"mean_token_accuracy": 0.12589406818151475,
|
|
"num_tokens": 3596785.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.048238277435303,
|
|
"epoch": 1.8294380905022378,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004905243956138298,
|
|
"loss": 6.7346,
|
|
"mean_token_accuracy": 0.135438484698534,
|
|
"num_tokens": 3605357.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.994794654846191,
|
|
"epoch": 1.834410740924913,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004904119312345896,
|
|
"loss": 6.7375,
|
|
"mean_token_accuracy": 0.13547348901629447,
|
|
"num_tokens": 3614979.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.0416075706481935,
|
|
"epoch": 1.8393833913475883,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004902988179038492,
|
|
"loss": 6.7421,
|
|
"mean_token_accuracy": 0.12939074113965035,
|
|
"num_tokens": 3624594.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.009415578842163,
|
|
"epoch": 1.8443560417702636,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000490185055962374,
|
|
"loss": 6.7407,
|
|
"mean_token_accuracy": 0.12785973995923997,
|
|
"num_tokens": 3633998.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.007656097412109,
|
|
"epoch": 1.8493286921929388,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004900706457528845,
|
|
"loss": 6.7177,
|
|
"mean_token_accuracy": 0.13221473544836043,
|
|
"num_tokens": 3642406.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.952472496032715,
|
|
"epoch": 1.854301342615614,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004899555876200535,
|
|
"loss": 6.7682,
|
|
"mean_token_accuracy": 0.1311837539076805,
|
|
"num_tokens": 3652905.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.976423645019532,
|
|
"epoch": 1.8592739930382893,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004898398819105059,
|
|
"loss": 6.7005,
|
|
"mean_token_accuracy": 0.13031949773430823,
|
|
"num_tokens": 3662518.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.999481439590454,
|
|
"epoch": 1.8642466434609646,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004897235289728178,
|
|
"loss": 6.7608,
|
|
"mean_token_accuracy": 0.13540611416101456,
|
|
"num_tokens": 3672102.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.021488952636719,
|
|
"epoch": 1.8692192938836398,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004896065291575145,
|
|
"loss": 6.7774,
|
|
"mean_token_accuracy": 0.12987095937132836,
|
|
"num_tokens": 3681821.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.1252607822418215,
|
|
"epoch": 1.8741919443063153,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004894888828170708,
|
|
"loss": 6.8562,
|
|
"mean_token_accuracy": 0.1253182515501976,
|
|
"num_tokens": 3690753.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.9700095653533936,
|
|
"epoch": 1.8791645947289906,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004893705903059087,
|
|
"loss": 6.8283,
|
|
"mean_token_accuracy": 0.1246264062821865,
|
|
"num_tokens": 3701694.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.992527627944947,
|
|
"epoch": 1.8841372451516658,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004892516519803971,
|
|
"loss": 6.719,
|
|
"mean_token_accuracy": 0.12994848415255547,
|
|
"num_tokens": 3710894.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.957933521270752,
|
|
"epoch": 1.8891098955743413,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004891320681988504,
|
|
"loss": 6.669,
|
|
"mean_token_accuracy": 0.13717258870601653,
|
|
"num_tokens": 3720081.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.985046005249023,
|
|
"epoch": 1.8940825459970165,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004890118393215277,
|
|
"loss": 6.7771,
|
|
"mean_token_accuracy": 0.1329424649477005,
|
|
"num_tokens": 3731168.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.964144134521485,
|
|
"epoch": 1.8990551964196918,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004888909657106311,
|
|
"loss": 6.5844,
|
|
"mean_token_accuracy": 0.1396918587386608,
|
|
"num_tokens": 3740533.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.901882982254028,
|
|
"epoch": 1.904027846842367,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004887694477303055,
|
|
"loss": 6.7628,
|
|
"mean_token_accuracy": 0.12639055252075196,
|
|
"num_tokens": 3750523.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.009959077835083,
|
|
"epoch": 1.9090004972650423,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004886472857466368,
|
|
"loss": 6.7959,
|
|
"mean_token_accuracy": 0.12752603068947793,
|
|
"num_tokens": 3760323.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.854344367980957,
|
|
"epoch": 1.9139731476877175,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004885244801276509,
|
|
"loss": 6.5538,
|
|
"mean_token_accuracy": 0.1511863052845001,
|
|
"num_tokens": 3770970.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.952729272842407,
|
|
"epoch": 1.9189457981103928,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004884010312433131,
|
|
"loss": 6.7821,
|
|
"mean_token_accuracy": 0.13067351803183555,
|
|
"num_tokens": 3780810.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.96802020072937,
|
|
"epoch": 1.923918448533068,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048827693946552615,
|
|
"loss": 6.6823,
|
|
"mean_token_accuracy": 0.13707536160945893,
|
|
"num_tokens": 3791461.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.8907379627227785,
|
|
"epoch": 1.9288910989557433,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048815220516813004,
|
|
"loss": 6.484,
|
|
"mean_token_accuracy": 0.14400224685668944,
|
|
"num_tokens": 3800353.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.949782991409302,
|
|
"epoch": 1.9338637493784185,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004880268287269,
|
|
"loss": 6.6703,
|
|
"mean_token_accuracy": 0.13870111033320426,
|
|
"num_tokens": 3810633.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 7.001215553283691,
|
|
"epoch": 1.938836399801094,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048790081051954605,
|
|
"loss": 6.8184,
|
|
"mean_token_accuracy": 0.12762571945786477,
|
|
"num_tokens": 3820254.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.051752328872681,
|
|
"epoch": 1.9438090502237693,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004877741509257116,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.1287897400557995,
|
|
"num_tokens": 3829976.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.898312187194824,
|
|
"epoch": 1.9487817006464445,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004876468503269721,
|
|
"loss": 6.7395,
|
|
"mean_token_accuracy": 0.12937736362218857,
|
|
"num_tokens": 3840593.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.989482784271241,
|
|
"epoch": 1.95375435106912,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048751890910683434,
|
|
"loss": 6.7072,
|
|
"mean_token_accuracy": 0.13608901649713517,
|
|
"num_tokens": 3850063.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.022893190383911,
|
|
"epoch": 1.9587270014917952,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048739032765073487,
|
|
"loss": 6.7606,
|
|
"mean_token_accuracy": 0.13344680294394493,
|
|
"num_tokens": 3860132.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.91834192276001,
|
|
"epoch": 1.9636996519144705,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048726110634603913,
|
|
"loss": 6.716,
|
|
"mean_token_accuracy": 0.13682736381888388,
|
|
"num_tokens": 3869914.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.905009174346924,
|
|
"epoch": 1.9686723023371457,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048713124558204013,
|
|
"loss": 6.6497,
|
|
"mean_token_accuracy": 0.13977741748094558,
|
|
"num_tokens": 3879825.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.998669290542603,
|
|
"epoch": 1.973644952759821,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004870007457499573,
|
|
"loss": 6.7237,
|
|
"mean_token_accuracy": 0.13763411790132524,
|
|
"num_tokens": 3889318.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.947281312942505,
|
|
"epoch": 1.9786176031824962,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048686960724293537,
|
|
"loss": 6.7166,
|
|
"mean_token_accuracy": 0.13478432670235635,
|
|
"num_tokens": 3899017.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.998880863189697,
|
|
"epoch": 1.9835902536051715,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004867378304560432,
|
|
"loss": 6.6623,
|
|
"mean_token_accuracy": 0.13629551306366922,
|
|
"num_tokens": 3907766.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.9843543529510494,
|
|
"epoch": 1.9885629040278467,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004866054157862723,
|
|
"loss": 6.6713,
|
|
"mean_token_accuracy": 0.1326850362122059,
|
|
"num_tokens": 3916323.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9885629040278467,
|
|
"eval_entropy": 6.683325477459464,
|
|
"eval_loss": 6.8779425621032715,
|
|
"eval_mean_token_accuracy": 0.12322797361881502,
|
|
"eval_num_tokens": 3916323.0,
|
|
"eval_runtime": 1.96,
|
|
"eval_samples_per_second": 1767.897,
|
|
"eval_steps_per_second": 221.434,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.87865195274353,
|
|
"epoch": 1.993535554450522,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004864723636325361,
|
|
"loss": 6.7465,
|
|
"mean_token_accuracy": 0.12982959896326066,
|
|
"num_tokens": 3927639.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 7.003038835525513,
|
|
"epoch": 1.9985082048731975,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00048633867439566855,
|
|
"loss": 6.6758,
|
|
"mean_token_accuracy": 0.12855788990855216,
|
|
"num_tokens": 3937452.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 7.019687546624078,
|
|
"epoch": 2.002983590253605,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004862043484784227,
|
|
"loss": 6.6319,
|
|
"mean_token_accuracy": 0.13641717616054747,
|
|
"num_tokens": 3946737.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.986137628555298,
|
|
"epoch": 2.0079562406762803,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048606938628546975,
|
|
"loss": 6.4448,
|
|
"mean_token_accuracy": 0.13515895754098892,
|
|
"num_tokens": 3957541.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.9740211963653564,
|
|
"epoch": 2.0129288910989556,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004859337882233978,
|
|
"loss": 6.5138,
|
|
"mean_token_accuracy": 0.1342613883316517,
|
|
"num_tokens": 3967621.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.984215354919433,
|
|
"epoch": 2.017901541521631,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004857975547007106,
|
|
"loss": 6.5171,
|
|
"mean_token_accuracy": 0.13852396756410598,
|
|
"num_tokens": 3977920.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.763665580749512,
|
|
"epoch": 2.0228741919443065,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048566068612782595,
|
|
"loss": 6.1818,
|
|
"mean_token_accuracy": 0.16204103827476501,
|
|
"num_tokens": 3988128.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.911743068695069,
|
|
"epoch": 2.0278468423669818,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048552318291707533,
|
|
"loss": 6.3774,
|
|
"mean_token_accuracy": 0.1404424302279949,
|
|
"num_tokens": 3998467.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.913117694854736,
|
|
"epoch": 2.032819492789657,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048538504548270177,
|
|
"loss": 6.3512,
|
|
"mean_token_accuracy": 0.13867299407720565,
|
|
"num_tokens": 4007513.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.781928539276123,
|
|
"epoch": 2.0377921432123323,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048524627424085913,
|
|
"loss": 6.3905,
|
|
"mean_token_accuracy": 0.14841719791293145,
|
|
"num_tokens": 4016538.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.858273601531982,
|
|
"epoch": 2.0427647936350075,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048510686960961057,
|
|
"loss": 6.3456,
|
|
"mean_token_accuracy": 0.14724784344434738,
|
|
"num_tokens": 4026850.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.8967320919036865,
|
|
"epoch": 2.0477374440576828,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004849668320089276,
|
|
"loss": 6.3173,
|
|
"mean_token_accuracy": 0.14462508261203766,
|
|
"num_tokens": 4037105.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.828446769714356,
|
|
"epoch": 2.052710094480358,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048482616186068826,
|
|
"loss": 6.3537,
|
|
"mean_token_accuracy": 0.14934856966137885,
|
|
"num_tokens": 4045694.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.767114162445068,
|
|
"epoch": 2.0576827449030333,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048468485958867664,
|
|
"loss": 6.2814,
|
|
"mean_token_accuracy": 0.14825050681829452,
|
|
"num_tokens": 4056425.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.773030090332031,
|
|
"epoch": 2.0626553953257085,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048454292561858083,
|
|
"loss": 6.288,
|
|
"mean_token_accuracy": 0.15561603903770446,
|
|
"num_tokens": 4066257.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.842053651809692,
|
|
"epoch": 2.0676280457483838,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004844003603779921,
|
|
"loss": 6.3713,
|
|
"mean_token_accuracy": 0.14501523673534394,
|
|
"num_tokens": 4076835.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.833949851989746,
|
|
"epoch": 2.072600696171059,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004842571642964036,
|
|
"loss": 6.4639,
|
|
"mean_token_accuracy": 0.14105072170495986,
|
|
"num_tokens": 4087644.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.86323914527893,
|
|
"epoch": 2.0775733465937343,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048411333780520876,
|
|
"loss": 6.4035,
|
|
"mean_token_accuracy": 0.14216212481260299,
|
|
"num_tokens": 4097040.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.7887468338012695,
|
|
"epoch": 2.0825459970164095,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048396888133770034,
|
|
"loss": 6.3848,
|
|
"mean_token_accuracy": 0.14581385552883147,
|
|
"num_tokens": 4106698.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.830855941772461,
|
|
"epoch": 2.087518647439085,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048382379532906897,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.14782741963863372,
|
|
"num_tokens": 4116804.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.877511978149414,
|
|
"epoch": 2.0924912978617605,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004836780802164016,
|
|
"loss": 6.4806,
|
|
"mean_token_accuracy": 0.13519393280148506,
|
|
"num_tokens": 4126658.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.7930272102355955,
|
|
"epoch": 2.0974639482844357,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048353173643868064,
|
|
"loss": 6.2905,
|
|
"mean_token_accuracy": 0.14840685427188874,
|
|
"num_tokens": 4136323.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.7793303489685055,
|
|
"epoch": 2.102436598707111,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004833847644367826,
|
|
"loss": 6.4032,
|
|
"mean_token_accuracy": 0.14725218415260316,
|
|
"num_tokens": 4145714.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.9044290542602536,
|
|
"epoch": 2.107409249129786,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00048323716465347615,
|
|
"loss": 6.3931,
|
|
"mean_token_accuracy": 0.1457737736403942,
|
|
"num_tokens": 4156059.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.804062080383301,
|
|
"epoch": 2.1123818995524615,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004830889375334215,
|
|
"loss": 6.4462,
|
|
"mean_token_accuracy": 0.13992414698004724,
|
|
"num_tokens": 4166990.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.808424854278565,
|
|
"epoch": 2.1173545499751367,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048294008352316875,
|
|
"loss": 6.3931,
|
|
"mean_token_accuracy": 0.14933390244841577,
|
|
"num_tokens": 4175716.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.769442796707153,
|
|
"epoch": 2.122327200397812,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004827906030711564,
|
|
"loss": 6.422,
|
|
"mean_token_accuracy": 0.13968386873602867,
|
|
"num_tokens": 4184832.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.654538011550903,
|
|
"epoch": 2.1272998508204872,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048264049662771053,
|
|
"loss": 6.207,
|
|
"mean_token_accuracy": 0.15818047374486924,
|
|
"num_tokens": 4195678.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.822854089736938,
|
|
"epoch": 2.1322725012431625,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048248976464504266,
|
|
"loss": 6.3875,
|
|
"mean_token_accuracy": 0.14510672092437743,
|
|
"num_tokens": 4205357.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.867270231246948,
|
|
"epoch": 2.1372451516658377,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048233840757724926,
|
|
"loss": 6.459,
|
|
"mean_token_accuracy": 0.1386936880648136,
|
|
"num_tokens": 4215416.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.840628004074096,
|
|
"epoch": 2.142217802088513,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048218642588030943,
|
|
"loss": 6.4929,
|
|
"mean_token_accuracy": 0.13991847559809684,
|
|
"num_tokens": 4225894.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.856436395645142,
|
|
"epoch": 2.1471904525111887,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048203382001208453,
|
|
"loss": 6.4219,
|
|
"mean_token_accuracy": 0.1450898639857769,
|
|
"num_tokens": 4235973.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.797693395614624,
|
|
"epoch": 2.152163102933864,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048188059043231585,
|
|
"loss": 6.3518,
|
|
"mean_token_accuracy": 0.14888587296009065,
|
|
"num_tokens": 4246411.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.86753396987915,
|
|
"epoch": 2.157135753356539,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048172673760262414,
|
|
"loss": 6.3836,
|
|
"mean_token_accuracy": 0.14080112874507905,
|
|
"num_tokens": 4255754.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.750627756118774,
|
|
"epoch": 2.1621084037792144,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048157226198650735,
|
|
"loss": 6.3364,
|
|
"mean_token_accuracy": 0.14296146482229233,
|
|
"num_tokens": 4265263.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.902114629745483,
|
|
"epoch": 2.1670810542018897,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004814171640493398,
|
|
"loss": 6.5105,
|
|
"mean_token_accuracy": 0.13793230652809144,
|
|
"num_tokens": 4275262.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.783820867538452,
|
|
"epoch": 2.172053704624565,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004812614442583708,
|
|
"loss": 6.4154,
|
|
"mean_token_accuracy": 0.1436000294983387,
|
|
"num_tokens": 4285467.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.776021480560303,
|
|
"epoch": 2.17702635504724,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004811051030827227,
|
|
"loss": 6.3484,
|
|
"mean_token_accuracy": 0.1466057628393173,
|
|
"num_tokens": 4294680.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.777132940292359,
|
|
"epoch": 2.1819990054699154,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048094814099339006,
|
|
"loss": 6.4251,
|
|
"mean_token_accuracy": 0.1400954969227314,
|
|
"num_tokens": 4304605.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.708856534957886,
|
|
"epoch": 2.1869716558925907,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004807905584632381,
|
|
"loss": 6.2508,
|
|
"mean_token_accuracy": 0.15585448890924453,
|
|
"num_tokens": 4314112.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.69225287437439,
|
|
"epoch": 2.191944306315266,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048063235596700087,
|
|
"loss": 6.3016,
|
|
"mean_token_accuracy": 0.14862677007913588,
|
|
"num_tokens": 4323872.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.80482668876648,
|
|
"epoch": 2.196916956737941,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048047353398128056,
|
|
"loss": 6.2885,
|
|
"mean_token_accuracy": 0.1484292671084404,
|
|
"num_tokens": 4333002.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.753137302398682,
|
|
"epoch": 2.2018896071606164,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004803140929845452,
|
|
"loss": 6.3902,
|
|
"mean_token_accuracy": 0.14406114667654038,
|
|
"num_tokens": 4342012.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.775022172927857,
|
|
"epoch": 2.2068622575832917,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048015403345712796,
|
|
"loss": 6.3388,
|
|
"mean_token_accuracy": 0.14243211150169371,
|
|
"num_tokens": 4351959.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.845752477645874,
|
|
"epoch": 2.2118349080059674,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004799933558812252,
|
|
"loss": 6.3294,
|
|
"mean_token_accuracy": 0.15065963119268416,
|
|
"num_tokens": 4361267.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.742046594619751,
|
|
"epoch": 2.2168075584286426,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047983206074089546,
|
|
"loss": 6.3583,
|
|
"mean_token_accuracy": 0.1481700599193573,
|
|
"num_tokens": 4370104.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.486421012878418,
|
|
"epoch": 2.221780208851318,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004796701485220576,
|
|
"loss": 6.1389,
|
|
"mean_token_accuracy": 0.17412543892860413,
|
|
"num_tokens": 4380724.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.6284302234649655,
|
|
"epoch": 2.226752859273993,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004795076197124893,
|
|
"loss": 6.2173,
|
|
"mean_token_accuracy": 0.161598302423954,
|
|
"num_tokens": 4390521.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.787254476547242,
|
|
"epoch": 2.2317255096966684,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047934447480182626,
|
|
"loss": 6.3687,
|
|
"mean_token_accuracy": 0.14581730514764785,
|
|
"num_tokens": 4400075.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.728147888183594,
|
|
"epoch": 2.2366981601193436,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004791807142815598,
|
|
"loss": 6.3312,
|
|
"mean_token_accuracy": 0.14047415405511857,
|
|
"num_tokens": 4409913.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.774549102783203,
|
|
"epoch": 2.241670810542019,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047901633864503614,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.1467915117740631,
|
|
"num_tokens": 4418551.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.714389085769653,
|
|
"epoch": 2.246643460964694,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047885134838745433,
|
|
"loss": 6.517,
|
|
"mean_token_accuracy": 0.14087366461753845,
|
|
"num_tokens": 4428129.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.912155628204346,
|
|
"epoch": 2.2516161113873694,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004786857440058654,
|
|
"loss": 6.4103,
|
|
"mean_token_accuracy": 0.14493027850985526,
|
|
"num_tokens": 4437311.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.607780694961548,
|
|
"epoch": 2.2565887618100446,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004785195259991701,
|
|
"loss": 6.2348,
|
|
"mean_token_accuracy": 0.15134743154048919,
|
|
"num_tokens": 4447387.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.6918535232543945,
|
|
"epoch": 2.26156141223272,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000478352694868118,
|
|
"loss": 6.4112,
|
|
"mean_token_accuracy": 0.14285444468259811,
|
|
"num_tokens": 4456435.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.658960962295533,
|
|
"epoch": 2.266534062655395,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004781852511153058,
|
|
"loss": 6.2484,
|
|
"mean_token_accuracy": 0.15608968287706376,
|
|
"num_tokens": 4466464.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.774130392074585,
|
|
"epoch": 2.271506713078071,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047801719524517573,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.1427334375679493,
|
|
"num_tokens": 4476793.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.7072478294372555,
|
|
"epoch": 2.276479363500746,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047784852776401403,
|
|
"loss": 6.3263,
|
|
"mean_token_accuracy": 0.15021902024745942,
|
|
"num_tokens": 4485811.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.743951034545899,
|
|
"epoch": 2.2814520139234213,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004776792491799495,
|
|
"loss": 6.4373,
|
|
"mean_token_accuracy": 0.1443065285682678,
|
|
"num_tokens": 4495018.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.772998905181884,
|
|
"epoch": 2.2864246643460966,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047750936000295213,
|
|
"loss": 6.4176,
|
|
"mean_token_accuracy": 0.14620297253131867,
|
|
"num_tokens": 4505847.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.702749156951905,
|
|
"epoch": 2.291397314768772,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047733886074483115,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.14058504924178122,
|
|
"num_tokens": 4516104.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.6128565788269045,
|
|
"epoch": 2.296369965191447,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004771677519192339,
|
|
"loss": 6.1727,
|
|
"mean_token_accuracy": 0.16449575871229172,
|
|
"num_tokens": 4524875.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.714238214492798,
|
|
"epoch": 2.3013426156141223,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004769960340416439,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.14073914885520936,
|
|
"num_tokens": 4533836.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.5153923511505125,
|
|
"epoch": 2.3063152660367976,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047682370762937976,
|
|
"loss": 5.9221,
|
|
"mean_token_accuracy": 0.1791136533021927,
|
|
"num_tokens": 4544323.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.724068832397461,
|
|
"epoch": 2.311287916459473,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047665077320159315,
|
|
"loss": 6.3449,
|
|
"mean_token_accuracy": 0.14938945770263673,
|
|
"num_tokens": 4553811.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.606375408172608,
|
|
"epoch": 2.316260566882148,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.00047647723127926765,
|
|
"loss": 6.2023,
|
|
"mean_token_accuracy": 0.16304922550916673,
|
|
"num_tokens": 4563098.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.750341415405273,
|
|
"epoch": 2.3212332173048233,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004763030823852168,
|
|
"loss": 6.4698,
|
|
"mean_token_accuracy": 0.14536633044481279,
|
|
"num_tokens": 4573589.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.775747060775757,
|
|
"epoch": 2.3262058677274986,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004761283270440829,
|
|
"loss": 6.376,
|
|
"mean_token_accuracy": 0.14753315448760987,
|
|
"num_tokens": 4583059.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.800399255752564,
|
|
"epoch": 2.331178518150174,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004759529657823349,
|
|
"loss": 6.5477,
|
|
"mean_token_accuracy": 0.1409124456346035,
|
|
"num_tokens": 4593371.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.790610027313233,
|
|
"epoch": 2.336151168572849,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00047577699912826755,
|
|
"loss": 6.4511,
|
|
"mean_token_accuracy": 0.1407085582613945,
|
|
"num_tokens": 4602862.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.7324306011199955,
|
|
"epoch": 2.341123818995525,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047560042761199915,
|
|
"loss": 6.3316,
|
|
"mean_token_accuracy": 0.14536074548959732,
|
|
"num_tokens": 4612827.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.701028776168823,
|
|
"epoch": 2.3460964694182,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004754232517654704,
|
|
"loss": 6.3802,
|
|
"mean_token_accuracy": 0.1461087703704834,
|
|
"num_tokens": 4621706.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.751887369155884,
|
|
"epoch": 2.3510691198408753,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004752454721224424,
|
|
"loss": 6.345,
|
|
"mean_token_accuracy": 0.1457712844014168,
|
|
"num_tokens": 4630878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.7952159404754635,
|
|
"epoch": 2.3560417702635505,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047506708921849535,
|
|
"loss": 6.4574,
|
|
"mean_token_accuracy": 0.13766919448971748,
|
|
"num_tokens": 4640582.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.6416051387786865,
|
|
"epoch": 2.361014420686226,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004748881035910269,
|
|
"loss": 6.3986,
|
|
"mean_token_accuracy": 0.1460815779864788,
|
|
"num_tokens": 4650799.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.777605295181274,
|
|
"epoch": 2.365987071108901,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004747085157792504,
|
|
"loss": 6.3271,
|
|
"mean_token_accuracy": 0.15237243473529816,
|
|
"num_tokens": 4660566.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.724439668655395,
|
|
"epoch": 2.3709597215315763,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004745283263241933,
|
|
"loss": 6.4313,
|
|
"mean_token_accuracy": 0.14020977020263672,
|
|
"num_tokens": 4670234.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.730980968475341,
|
|
"epoch": 2.3759323719542516,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004743475357686957,
|
|
"loss": 6.3769,
|
|
"mean_token_accuracy": 0.1434305027127266,
|
|
"num_tokens": 4681114.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.678455448150634,
|
|
"epoch": 2.380905022376927,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047416614465740843,
|
|
"loss": 6.1761,
|
|
"mean_token_accuracy": 0.16967258751392364,
|
|
"num_tokens": 4691970.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.647848844528198,
|
|
"epoch": 2.385877672799602,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047398415353679167,
|
|
"loss": 6.3655,
|
|
"mean_token_accuracy": 0.14517362117767335,
|
|
"num_tokens": 4701723.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.644217252731323,
|
|
"epoch": 2.3908503232222773,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004738015629551132,
|
|
"loss": 6.3524,
|
|
"mean_token_accuracy": 0.14786965325474738,
|
|
"num_tokens": 4711356.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.731314420700073,
|
|
"epoch": 2.395822973644953,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047361837346244655,
|
|
"loss": 6.3752,
|
|
"mean_token_accuracy": 0.15293737798929213,
|
|
"num_tokens": 4721765.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.701357460021972,
|
|
"epoch": 2.4007956240676283,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004734345856106698,
|
|
"loss": 6.2699,
|
|
"mean_token_accuracy": 0.14782674014568328,
|
|
"num_tokens": 4731678.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.6664855003356935,
|
|
"epoch": 2.4057682744903035,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047325019995346355,
|
|
"loss": 6.2902,
|
|
"mean_token_accuracy": 0.1536775067448616,
|
|
"num_tokens": 4740960.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.663874244689941,
|
|
"epoch": 2.4107409249129788,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004730652170463091,
|
|
"loss": 6.4179,
|
|
"mean_token_accuracy": 0.13955927267670631,
|
|
"num_tokens": 4751753.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.706752920150757,
|
|
"epoch": 2.415713575335654,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004728796374464875,
|
|
"loss": 6.2962,
|
|
"mean_token_accuracy": 0.15075381994247436,
|
|
"num_tokens": 4761274.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.594811296463012,
|
|
"epoch": 2.4206862257583293,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047269346171307716,
|
|
"loss": 6.4412,
|
|
"mean_token_accuracy": 0.14638966023921968,
|
|
"num_tokens": 4770852.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.853709268569946,
|
|
"epoch": 2.4256588761810045,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004725066904069523,
|
|
"loss": 6.3163,
|
|
"mean_token_accuracy": 0.14985065162181854,
|
|
"num_tokens": 4780857.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.611556529998779,
|
|
"epoch": 2.4306315266036798,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004723193240907814,
|
|
"loss": 6.2613,
|
|
"mean_token_accuracy": 0.14320006519556044,
|
|
"num_tokens": 4790018.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.631385469436646,
|
|
"epoch": 2.435604177026355,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004721313633290257,
|
|
"loss": 6.3602,
|
|
"mean_token_accuracy": 0.15065304413437844,
|
|
"num_tokens": 4799747.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.80011568069458,
|
|
"epoch": 2.4405768274490303,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000471942808687937,
|
|
"loss": 6.3755,
|
|
"mean_token_accuracy": 0.14567220509052276,
|
|
"num_tokens": 4809710.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.701170015335083,
|
|
"epoch": 2.4455494778717055,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047175366073555654,
|
|
"loss": 6.3245,
|
|
"mean_token_accuracy": 0.15504877418279647,
|
|
"num_tokens": 4819387.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.6124413967132565,
|
|
"epoch": 2.4505221282943808,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004715639200417125,
|
|
"loss": 6.2999,
|
|
"mean_token_accuracy": 0.147188264131546,
|
|
"num_tokens": 4829541.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.65046272277832,
|
|
"epoch": 2.455494778717056,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004713735871780192,
|
|
"loss": 6.2869,
|
|
"mean_token_accuracy": 0.1503270223736763,
|
|
"num_tokens": 4838699.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.691880512237549,
|
|
"epoch": 2.4604674291397313,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004711826627178747,
|
|
"loss": 6.4055,
|
|
"mean_token_accuracy": 0.14789262041449547,
|
|
"num_tokens": 4848295.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.639479637145996,
|
|
"epoch": 2.4654400795624065,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047099114723645944,
|
|
"loss": 6.2821,
|
|
"mean_token_accuracy": 0.15444958806037903,
|
|
"num_tokens": 4857649.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.69353175163269,
|
|
"epoch": 2.470412729985082,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047079904131073434,
|
|
"loss": 6.2909,
|
|
"mean_token_accuracy": 0.1514730140566826,
|
|
"num_tokens": 4867332.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.664346742630005,
|
|
"epoch": 2.4753853804077575,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004706063455194389,
|
|
"loss": 6.3655,
|
|
"mean_token_accuracy": 0.13988956362009047,
|
|
"num_tokens": 4878726.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.740277051925659,
|
|
"epoch": 2.4803580308304327,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047041306044308986,
|
|
"loss": 6.3808,
|
|
"mean_token_accuracy": 0.1416482426226139,
|
|
"num_tokens": 4888030.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.679982471466064,
|
|
"epoch": 2.485330681253108,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004702191866639792,
|
|
"loss": 6.3469,
|
|
"mean_token_accuracy": 0.14078416377305986,
|
|
"num_tokens": 4897191.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.485330681253108,
|
|
"eval_entropy": 6.521233376270065,
|
|
"eval_loss": 6.747542381286621,
|
|
"eval_mean_token_accuracy": 0.13297974515689134,
|
|
"eval_num_tokens": 4897191.0,
|
|
"eval_runtime": 1.9427,
|
|
"eval_samples_per_second": 1783.581,
|
|
"eval_steps_per_second": 223.398,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.694563913345337,
|
|
"epoch": 2.490303331675783,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004700247247661726,
|
|
"loss": 6.3442,
|
|
"mean_token_accuracy": 0.15119849890470505,
|
|
"num_tokens": 4907154.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.492526578903198,
|
|
"epoch": 2.4952759820984585,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004698296753355072,
|
|
"loss": 6.0197,
|
|
"mean_token_accuracy": 0.17371601834893227,
|
|
"num_tokens": 4917582.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.595386600494384,
|
|
"epoch": 2.5002486325211337,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004696340389595903,
|
|
"loss": 6.2863,
|
|
"mean_token_accuracy": 0.15509023070335387,
|
|
"num_tokens": 4927473.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.724164581298828,
|
|
"epoch": 2.505221282943809,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046943781622779733,
|
|
"loss": 6.4139,
|
|
"mean_token_accuracy": 0.13926264271140099,
|
|
"num_tokens": 4938037.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.658544635772705,
|
|
"epoch": 2.510193933366484,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046924100773127055,
|
|
"loss": 6.3772,
|
|
"mean_token_accuracy": 0.14309081956744193,
|
|
"num_tokens": 4948009.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.619173192977906,
|
|
"epoch": 2.5151665837891595,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046904361406291633,
|
|
"loss": 6.3617,
|
|
"mean_token_accuracy": 0.14876620322465897,
|
|
"num_tokens": 4957971.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.760876417160034,
|
|
"epoch": 2.520139234211835,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004688456358174044,
|
|
"loss": 6.344,
|
|
"mean_token_accuracy": 0.1419855050742626,
|
|
"num_tokens": 4968040.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.6025079727172855,
|
|
"epoch": 2.5251118846345104,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004686470735911653,
|
|
"loss": 6.269,
|
|
"mean_token_accuracy": 0.15409234911203384,
|
|
"num_tokens": 4977142.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.636661720275879,
|
|
"epoch": 2.5300845350571857,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004684479279823891,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.1451442778110504,
|
|
"num_tokens": 4986726.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.798409509658813,
|
|
"epoch": 2.535057185479861,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004682481995910232,
|
|
"loss": 6.3353,
|
|
"mean_token_accuracy": 0.1390838772058487,
|
|
"num_tokens": 4996690.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.639165019989013,
|
|
"epoch": 2.540029835902536,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046804788901877084,
|
|
"loss": 6.3825,
|
|
"mean_token_accuracy": 0.14859820157289505,
|
|
"num_tokens": 5005804.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.618333339691162,
|
|
"epoch": 2.5450024863252114,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00046784699686908886,
|
|
"loss": 6.3234,
|
|
"mean_token_accuracy": 0.15142451822757722,
|
|
"num_tokens": 5015666.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.639498615264893,
|
|
"epoch": 2.5499751367478867,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00046764552374718656,
|
|
"loss": 6.2638,
|
|
"mean_token_accuracy": 0.15632327049970626,
|
|
"num_tokens": 5025897.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.625174283981323,
|
|
"epoch": 2.554947787170562,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046744347026002315,
|
|
"loss": 6.3327,
|
|
"mean_token_accuracy": 0.15196211785078048,
|
|
"num_tokens": 5035195.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.658428955078125,
|
|
"epoch": 2.559920437593237,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004672408370163066,
|
|
"loss": 6.3407,
|
|
"mean_token_accuracy": 0.15045792758464813,
|
|
"num_tokens": 5044472.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.65118727684021,
|
|
"epoch": 2.5648930880159124,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000467037624626491,
|
|
"loss": 6.3786,
|
|
"mean_token_accuracy": 0.14528095722198486,
|
|
"num_tokens": 5053593.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.650417184829712,
|
|
"epoch": 2.5698657384385877,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046683383370277554,
|
|
"loss": 6.3683,
|
|
"mean_token_accuracy": 0.14565744251012802,
|
|
"num_tokens": 5063604.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.632540607452393,
|
|
"epoch": 2.574838388861263,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046662946485910226,
|
|
"loss": 6.2481,
|
|
"mean_token_accuracy": 0.14790819436311722,
|
|
"num_tokens": 5073013.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.625475215911865,
|
|
"epoch": 2.579811039283938,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000466424518711154,
|
|
"loss": 6.3418,
|
|
"mean_token_accuracy": 0.1495043784379959,
|
|
"num_tokens": 5082590.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.636396598815918,
|
|
"epoch": 2.5847836897066134,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004662189958763531,
|
|
"loss": 6.3149,
|
|
"mean_token_accuracy": 0.1487736500799656,
|
|
"num_tokens": 5091860.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.67173204421997,
|
|
"epoch": 2.5897563401292887,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004660128969738591,
|
|
"loss": 6.2393,
|
|
"mean_token_accuracy": 0.15700285136699677,
|
|
"num_tokens": 5101164.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.566663312911987,
|
|
"epoch": 2.594728990551964,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004658062226245668,
|
|
"loss": 6.337,
|
|
"mean_token_accuracy": 0.14751999974250793,
|
|
"num_tokens": 5112885.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.64238486289978,
|
|
"epoch": 2.5997016409746396,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004655989734511052,
|
|
"loss": 6.3003,
|
|
"mean_token_accuracy": 0.15081872791051865,
|
|
"num_tokens": 5124063.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.536421632766723,
|
|
"epoch": 2.604674291397315,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004653911500778343,
|
|
"loss": 6.1345,
|
|
"mean_token_accuracy": 0.16544700413942337,
|
|
"num_tokens": 5134286.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.5944748878479,
|
|
"epoch": 2.60964694181999,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004651827531308443,
|
|
"loss": 6.3869,
|
|
"mean_token_accuracy": 0.14455192983150483,
|
|
"num_tokens": 5144421.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.756843376159668,
|
|
"epoch": 2.6146195922426654,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004649737832379534,
|
|
"loss": 6.3407,
|
|
"mean_token_accuracy": 0.14513533413410187,
|
|
"num_tokens": 5153466.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.606828689575195,
|
|
"epoch": 2.6195922426653406,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004647642410287057,
|
|
"loss": 6.4036,
|
|
"mean_token_accuracy": 0.14541415348649026,
|
|
"num_tokens": 5164034.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.652927494049072,
|
|
"epoch": 2.624564893088016,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004645541271343696,
|
|
"loss": 6.3294,
|
|
"mean_token_accuracy": 0.1516832232475281,
|
|
"num_tokens": 5173867.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.621406412124633,
|
|
"epoch": 2.629537543510691,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046434344218793566,
|
|
"loss": 6.3767,
|
|
"mean_token_accuracy": 0.14626053124666213,
|
|
"num_tokens": 5183182.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.624752521514893,
|
|
"epoch": 2.6345101939333664,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046413218682411485,
|
|
"loss": 6.3335,
|
|
"mean_token_accuracy": 0.1461806558072567,
|
|
"num_tokens": 5192537.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.700682973861694,
|
|
"epoch": 2.6394828443560416,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004639203616793365,
|
|
"loss": 6.2993,
|
|
"mean_token_accuracy": 0.14982734620571136,
|
|
"num_tokens": 5201879.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.656138801574707,
|
|
"epoch": 2.6444554947787173,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046370796739174657,
|
|
"loss": 6.3236,
|
|
"mean_token_accuracy": 0.14249756708741187,
|
|
"num_tokens": 5211555.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.571947193145752,
|
|
"epoch": 2.6494281452013926,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046349500460120556,
|
|
"loss": 6.227,
|
|
"mean_token_accuracy": 0.1545217826962471,
|
|
"num_tokens": 5220895.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.681439971923828,
|
|
"epoch": 2.654400795624068,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046328147394928684,
|
|
"loss": 6.356,
|
|
"mean_token_accuracy": 0.15007806271314622,
|
|
"num_tokens": 5230101.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.718551921844482,
|
|
"epoch": 2.659373446046743,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004630673760792741,
|
|
"loss": 6.36,
|
|
"mean_token_accuracy": 0.13901026099920272,
|
|
"num_tokens": 5239472.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.590527725219727,
|
|
"epoch": 2.6643460964694183,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004628527116361603,
|
|
"loss": 6.2988,
|
|
"mean_token_accuracy": 0.14799102395772934,
|
|
"num_tokens": 5248823.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.628971290588379,
|
|
"epoch": 2.6693187468920936,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004626374812666449,
|
|
"loss": 6.3111,
|
|
"mean_token_accuracy": 0.14377440363168717,
|
|
"num_tokens": 5258551.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.590078401565552,
|
|
"epoch": 2.674291397314769,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046242168561913256,
|
|
"loss": 6.1483,
|
|
"mean_token_accuracy": 0.1630126193165779,
|
|
"num_tokens": 5267933.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.731613731384277,
|
|
"epoch": 2.679264047737444,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046220532534373067,
|
|
"loss": 6.3666,
|
|
"mean_token_accuracy": 0.14824986532330514,
|
|
"num_tokens": 5277856.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.6385619163513185,
|
|
"epoch": 2.6842366981601193,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004619884010922477,
|
|
"loss": 6.3558,
|
|
"mean_token_accuracy": 0.14487922340631484,
|
|
"num_tokens": 5288127.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.664433431625366,
|
|
"epoch": 2.6892093485827946,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004617709135181913,
|
|
"loss": 6.3048,
|
|
"mean_token_accuracy": 0.1482628807425499,
|
|
"num_tokens": 5297143.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.614977121353149,
|
|
"epoch": 2.69418199900547,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046155286327676594,
|
|
"loss": 6.2867,
|
|
"mean_token_accuracy": 0.15043259859085084,
|
|
"num_tokens": 5307067.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.586661243438721,
|
|
"epoch": 2.699154649428145,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00046133425102487134,
|
|
"loss": 6.2573,
|
|
"mean_token_accuracy": 0.1495133936405182,
|
|
"num_tokens": 5317471.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.699815559387207,
|
|
"epoch": 2.7041272998508203,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046111507742110035,
|
|
"loss": 6.3091,
|
|
"mean_token_accuracy": 0.15538086593151093,
|
|
"num_tokens": 5326803.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.596299743652343,
|
|
"epoch": 2.7090999502734956,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000460895343125737,
|
|
"loss": 6.3015,
|
|
"mean_token_accuracy": 0.15092724114656447,
|
|
"num_tokens": 5336342.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.508804607391357,
|
|
"epoch": 2.714072600696171,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046067504880075425,
|
|
"loss": 6.1327,
|
|
"mean_token_accuracy": 0.16722403913736344,
|
|
"num_tokens": 5345694.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.621329641342163,
|
|
"epoch": 2.719045251118846,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004604541951098122,
|
|
"loss": 6.2302,
|
|
"mean_token_accuracy": 0.15404069572687148,
|
|
"num_tokens": 5355277.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.585741662979126,
|
|
"epoch": 2.724017901541522,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004602327827182565,
|
|
"loss": 6.2384,
|
|
"mean_token_accuracy": 0.15714289322495462,
|
|
"num_tokens": 5364650.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.645495986938476,
|
|
"epoch": 2.728990551964197,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004600108122931156,
|
|
"loss": 6.2162,
|
|
"mean_token_accuracy": 0.15356728583574294,
|
|
"num_tokens": 5373151.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.571676301956177,
|
|
"epoch": 2.7339632023868723,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045978828450309907,
|
|
"loss": 6.3572,
|
|
"mean_token_accuracy": 0.14834558889269828,
|
|
"num_tokens": 5383093.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.69737548828125,
|
|
"epoch": 2.7389358528095475,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0004595652000185958,
|
|
"loss": 6.2549,
|
|
"mean_token_accuracy": 0.15186045691370964,
|
|
"num_tokens": 5392060.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.648882579803467,
|
|
"epoch": 2.743908503232223,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045934155951167163,
|
|
"loss": 6.3622,
|
|
"mean_token_accuracy": 0.1475387006998062,
|
|
"num_tokens": 5402119.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.640841960906982,
|
|
"epoch": 2.748881153654898,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004591173636560675,
|
|
"loss": 6.2705,
|
|
"mean_token_accuracy": 0.15139891654253007,
|
|
"num_tokens": 5411804.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.597275590896606,
|
|
"epoch": 2.7538538040775733,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004588926131271974,
|
|
"loss": 6.3379,
|
|
"mean_token_accuracy": 0.14531828910112382,
|
|
"num_tokens": 5423808.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.62809944152832,
|
|
"epoch": 2.7588264545002485,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045866730860214655,
|
|
"loss": 6.2968,
|
|
"mean_token_accuracy": 0.15449204593896865,
|
|
"num_tokens": 5432154.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.472979974746704,
|
|
"epoch": 2.763799104922924,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004584414507596687,
|
|
"loss": 6.0376,
|
|
"mean_token_accuracy": 0.1696406900882721,
|
|
"num_tokens": 5443006.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.517356491088867,
|
|
"epoch": 2.768771755345599,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045821504028018494,
|
|
"loss": 6.2237,
|
|
"mean_token_accuracy": 0.15524712949991226,
|
|
"num_tokens": 5453596.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.640099000930786,
|
|
"epoch": 2.7737444057682747,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004579880778457811,
|
|
"loss": 6.3266,
|
|
"mean_token_accuracy": 0.14540622234344483,
|
|
"num_tokens": 5463048.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.6180473327636715,
|
|
"epoch": 2.77871705619095,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004577605641402056,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.15415629595518113,
|
|
"num_tokens": 5472409.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.592954683303833,
|
|
"epoch": 2.7836897066136252,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004575324998488681,
|
|
"loss": 6.2083,
|
|
"mean_token_accuracy": 0.15716410130262376,
|
|
"num_tokens": 5481380.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.6298425674438475,
|
|
"epoch": 2.7886623570363005,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004573038856588364,
|
|
"loss": 6.4002,
|
|
"mean_token_accuracy": 0.14382093846797944,
|
|
"num_tokens": 5491157.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.532769441604614,
|
|
"epoch": 2.7936350074589757,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004570747222588354,
|
|
"loss": 6.3051,
|
|
"mean_token_accuracy": 0.1482105389237404,
|
|
"num_tokens": 5500856.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.603239250183106,
|
|
"epoch": 2.798607657881651,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045684501033924443,
|
|
"loss": 6.1989,
|
|
"mean_token_accuracy": 0.1541078582406044,
|
|
"num_tokens": 5509654.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.5999048233032225,
|
|
"epoch": 2.8035803083043263,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045661475059209503,
|
|
"loss": 6.1922,
|
|
"mean_token_accuracy": 0.156157885491848,
|
|
"num_tokens": 5519009.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.553065156936645,
|
|
"epoch": 2.8085529587270015,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004563839437110695,
|
|
"loss": 6.1993,
|
|
"mean_token_accuracy": 0.15775296539068223,
|
|
"num_tokens": 5528316.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.592592048645019,
|
|
"epoch": 2.8135256091496768,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045615259039149826,
|
|
"loss": 6.3304,
|
|
"mean_token_accuracy": 0.15206963419914246,
|
|
"num_tokens": 5538416.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.584443187713623,
|
|
"epoch": 2.818498259572352,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00045592069133035783,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.15321145057678223,
|
|
"num_tokens": 5548600.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.604082870483398,
|
|
"epoch": 2.8234709099950273,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004556882472262691,
|
|
"loss": 6.3412,
|
|
"mean_token_accuracy": 0.14708501547574998,
|
|
"num_tokens": 5559454.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.627229166030884,
|
|
"epoch": 2.8284435604177025,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004554552587794948,
|
|
"loss": 6.3036,
|
|
"mean_token_accuracy": 0.145070943236351,
|
|
"num_tokens": 5567748.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.656732416152954,
|
|
"epoch": 2.8334162108403778,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004552217266919375,
|
|
"loss": 6.3906,
|
|
"mean_token_accuracy": 0.144349205493927,
|
|
"num_tokens": 5577617.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.62886643409729,
|
|
"epoch": 2.838388861263053,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004549876516671376,
|
|
"loss": 6.3424,
|
|
"mean_token_accuracy": 0.14952508583664895,
|
|
"num_tokens": 5586600.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.627637052536011,
|
|
"epoch": 2.8433615116857283,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045475303441027114,
|
|
"loss": 6.2069,
|
|
"mean_token_accuracy": 0.1609795868396759,
|
|
"num_tokens": 5596526.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.580606412887573,
|
|
"epoch": 2.8483341621084035,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004545178756281479,
|
|
"loss": 6.2386,
|
|
"mean_token_accuracy": 0.15840020924806594,
|
|
"num_tokens": 5605780.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.511898469924927,
|
|
"epoch": 2.853306812531079,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00045428217602920876,
|
|
"loss": 6.2906,
|
|
"mean_token_accuracy": 0.15046771615743637,
|
|
"num_tokens": 5616109.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.534842681884766,
|
|
"epoch": 2.8582794629537545,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045404593632352385,
|
|
"loss": 6.2571,
|
|
"mean_token_accuracy": 0.15704997926950454,
|
|
"num_tokens": 5625423.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.680988407135009,
|
|
"epoch": 2.8632521133764297,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004538091572227909,
|
|
"loss": 6.3423,
|
|
"mean_token_accuracy": 0.1516302354633808,
|
|
"num_tokens": 5634538.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.525880241394043,
|
|
"epoch": 2.868224763799105,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004535718394403319,
|
|
"loss": 6.1145,
|
|
"mean_token_accuracy": 0.16714052483439445,
|
|
"num_tokens": 5644087.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.480349826812744,
|
|
"epoch": 2.87319741422178,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004533339836910924,
|
|
"loss": 6.1371,
|
|
"mean_token_accuracy": 0.1678331106901169,
|
|
"num_tokens": 5654241.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.588527631759644,
|
|
"epoch": 2.8781700646444555,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004530955906916382,
|
|
"loss": 6.2035,
|
|
"mean_token_accuracy": 0.1595718026161194,
|
|
"num_tokens": 5663827.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.623732614517212,
|
|
"epoch": 2.8831427150671307,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004528566611601536,
|
|
"loss": 6.2836,
|
|
"mean_token_accuracy": 0.14805630147457122,
|
|
"num_tokens": 5673427.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.588187885284424,
|
|
"epoch": 2.888115365489806,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004526171958164396,
|
|
"loss": 6.1662,
|
|
"mean_token_accuracy": 0.15485352501273156,
|
|
"num_tokens": 5684762.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.441930294036865,
|
|
"epoch": 2.893088015912481,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045237719538191107,
|
|
"loss": 6.2644,
|
|
"mean_token_accuracy": 0.15272388011217117,
|
|
"num_tokens": 5694938.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.583103132247925,
|
|
"epoch": 2.898060666335157,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000452136660579595,
|
|
"loss": 6.1903,
|
|
"mean_token_accuracy": 0.15786472707986832,
|
|
"num_tokens": 5704910.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.5423189163208,
|
|
"epoch": 2.903033316757832,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045189559213412825,
|
|
"loss": 6.2703,
|
|
"mean_token_accuracy": 0.14946384876966476,
|
|
"num_tokens": 5714325.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.539729833602905,
|
|
"epoch": 2.9080059671805074,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004516539907717553,
|
|
"loss": 6.1815,
|
|
"mean_token_accuracy": 0.161780047416687,
|
|
"num_tokens": 5724382.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.530433177947998,
|
|
"epoch": 2.9129786176031827,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045141185722032617,
|
|
"loss": 6.2304,
|
|
"mean_token_accuracy": 0.15309611409902574,
|
|
"num_tokens": 5734277.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.595566892623902,
|
|
"epoch": 2.917951268025858,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045116919220929407,
|
|
"loss": 6.2416,
|
|
"mean_token_accuracy": 0.14548557326197625,
|
|
"num_tokens": 5742772.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.542143440246582,
|
|
"epoch": 2.922923918448533,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004509259964697133,
|
|
"loss": 6.2609,
|
|
"mean_token_accuracy": 0.1522253632545471,
|
|
"num_tokens": 5754011.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.625977420806885,
|
|
"epoch": 2.9278965688712084,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000450682270734237,
|
|
"loss": 6.3417,
|
|
"mean_token_accuracy": 0.1489579387009144,
|
|
"num_tokens": 5765445.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.588511276245117,
|
|
"epoch": 2.9328692192938837,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004504380157371152,
|
|
"loss": 6.2587,
|
|
"mean_token_accuracy": 0.1521792843937874,
|
|
"num_tokens": 5775529.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.594973754882813,
|
|
"epoch": 2.937841869716559,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045019323221419204,
|
|
"loss": 6.4056,
|
|
"mean_token_accuracy": 0.14550215303897857,
|
|
"num_tokens": 5785594.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.552459049224853,
|
|
"epoch": 2.942814520139234,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044994792090290425,
|
|
"loss": 6.1165,
|
|
"mean_token_accuracy": 0.16292940378189086,
|
|
"num_tokens": 5796071.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.48425669670105,
|
|
"epoch": 2.9477871705619094,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044970208254227836,
|
|
"loss": 6.2307,
|
|
"mean_token_accuracy": 0.1594185084104538,
|
|
"num_tokens": 5805127.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.601228141784668,
|
|
"epoch": 2.9527598209845847,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044945571787292866,
|
|
"loss": 6.3117,
|
|
"mean_token_accuracy": 0.15017432123422622,
|
|
"num_tokens": 5815001.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.614652538299561,
|
|
"epoch": 2.95773247140726,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004492088276370552,
|
|
"loss": 6.2653,
|
|
"mean_token_accuracy": 0.1477195680141449,
|
|
"num_tokens": 5826915.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.588264083862304,
|
|
"epoch": 2.962705121829935,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004489614125784411,
|
|
"loss": 6.2603,
|
|
"mean_token_accuracy": 0.14955487325787545,
|
|
"num_tokens": 5837444.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.568655967712402,
|
|
"epoch": 2.9676777722526104,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004487134734424509,
|
|
"loss": 6.1253,
|
|
"mean_token_accuracy": 0.15760306864976883,
|
|
"num_tokens": 5846514.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.549760627746582,
|
|
"epoch": 2.9726504226752857,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004484650109760276,
|
|
"loss": 6.3821,
|
|
"mean_token_accuracy": 0.14270757064223288,
|
|
"num_tokens": 5857227.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.600679111480713,
|
|
"epoch": 2.9776230730979614,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004482160259276911,
|
|
"loss": 6.3714,
|
|
"mean_token_accuracy": 0.14973211139440537,
|
|
"num_tokens": 5866877.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.613196945190429,
|
|
"epoch": 2.9825957235206366,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004479665190475355,
|
|
"loss": 6.2483,
|
|
"mean_token_accuracy": 0.1548510044813156,
|
|
"num_tokens": 5877127.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.9825957235206366,
|
|
"eval_entropy": 6.347676026656331,
|
|
"eval_loss": 6.624464511871338,
|
|
"eval_mean_token_accuracy": 0.1370609762911011,
|
|
"eval_num_tokens": 5877127.0,
|
|
"eval_runtime": 1.9392,
|
|
"eval_samples_per_second": 1786.831,
|
|
"eval_steps_per_second": 223.805,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.478391075134278,
|
|
"epoch": 2.987568373943312,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044771649108722695,
|
|
"loss": 6.272,
|
|
"mean_token_accuracy": 0.15030792802572251,
|
|
"num_tokens": 5886434.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.547763872146606,
|
|
"epoch": 2.992541024365987,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004474659428000014,
|
|
"loss": 6.2691,
|
|
"mean_token_accuracy": 0.15371823012828828,
|
|
"num_tokens": 5895462.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.6339030265808105,
|
|
"epoch": 2.9975136747886624,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044721487494066254,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.15364308953285216,
|
|
"num_tokens": 5905274.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.461676491631402,
|
|
"epoch": 3.00198906016907,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000446963288265579,
|
|
"loss": 5.9883,
|
|
"mean_token_accuracy": 0.16431198683049944,
|
|
"num_tokens": 5914164.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.575624227523804,
|
|
"epoch": 3.0069617105917454,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00044671118353268257,
|
|
"loss": 5.8747,
|
|
"mean_token_accuracy": 0.16463825851678848,
|
|
"num_tokens": 5923159.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.497607851028443,
|
|
"epoch": 3.0119343610144207,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044645856150146593,
|
|
"loss": 5.8449,
|
|
"mean_token_accuracy": 0.16996805965900422,
|
|
"num_tokens": 5932708.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.547623920440674,
|
|
"epoch": 3.016907011437096,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044620542293297974,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.16645532250404357,
|
|
"num_tokens": 5941370.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.474026870727539,
|
|
"epoch": 3.021879661859771,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004459517685898312,
|
|
"loss": 5.8988,
|
|
"mean_token_accuracy": 0.1775730296969414,
|
|
"num_tokens": 5950567.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.546600770950318,
|
|
"epoch": 3.0268523122824464,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044569759923618113,
|
|
"loss": 5.9743,
|
|
"mean_token_accuracy": 0.16607621908187867,
|
|
"num_tokens": 5959242.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.534537076950073,
|
|
"epoch": 3.0318249627051217,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004454429156377419,
|
|
"loss": 5.9692,
|
|
"mean_token_accuracy": 0.1638436421751976,
|
|
"num_tokens": 5968365.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.503127765655518,
|
|
"epoch": 3.036797613127797,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004451877185617751,
|
|
"loss": 5.9452,
|
|
"mean_token_accuracy": 0.16334739774465562,
|
|
"num_tokens": 5978407.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.369291019439697,
|
|
"epoch": 3.041770263550472,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004449320087770894,
|
|
"loss": 5.6936,
|
|
"mean_token_accuracy": 0.19242739528417588,
|
|
"num_tokens": 5988865.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.460866975784302,
|
|
"epoch": 3.046742913973148,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004446757870540376,
|
|
"loss": 5.869,
|
|
"mean_token_accuracy": 0.17347271218895913,
|
|
"num_tokens": 5998873.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.4673107147216795,
|
|
"epoch": 3.051715564395823,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004444190541645153,
|
|
"loss": 5.8974,
|
|
"mean_token_accuracy": 0.16633763015270234,
|
|
"num_tokens": 6007021.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.402094125747681,
|
|
"epoch": 3.0566882148184984,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00044416181088195776,
|
|
"loss": 5.708,
|
|
"mean_token_accuracy": 0.17730897814035415,
|
|
"num_tokens": 6015884.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.407754755020141,
|
|
"epoch": 3.0616608652411736,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004439040579813379,
|
|
"loss": 5.9707,
|
|
"mean_token_accuracy": 0.1645668312907219,
|
|
"num_tokens": 6026110.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.268219089508056,
|
|
"epoch": 3.066633515663849,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00044364579623916385,
|
|
"loss": 5.708,
|
|
"mean_token_accuracy": 0.190046925842762,
|
|
"num_tokens": 6037479.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.497198390960693,
|
|
"epoch": 3.071606166086524,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004433870264334767,
|
|
"loss": 5.884,
|
|
"mean_token_accuracy": 0.16622215658426284,
|
|
"num_tokens": 6046743.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.408089208602905,
|
|
"epoch": 3.0765788165091994,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004431277493438484,
|
|
"loss": 5.9716,
|
|
"mean_token_accuracy": 0.16973157674074174,
|
|
"num_tokens": 6056982.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.481922006607055,
|
|
"epoch": 3.0815514669318746,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004428679657513789,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.1723087787628174,
|
|
"num_tokens": 6066016.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.3932819843292235,
|
|
"epoch": 3.08652411735455,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004426076764386939,
|
|
"loss": 5.8811,
|
|
"mean_token_accuracy": 0.15916473269462586,
|
|
"num_tokens": 6075149.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.472731590270996,
|
|
"epoch": 3.091496767777225,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00044234688218994293,
|
|
"loss": 5.8667,
|
|
"mean_token_accuracy": 0.1672999694943428,
|
|
"num_tokens": 6085736.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.389721345901489,
|
|
"epoch": 3.0964694181999004,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004420855837907967,
|
|
"loss": 5.7717,
|
|
"mean_token_accuracy": 0.18188717514276503,
|
|
"num_tokens": 6096557.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.533287906646729,
|
|
"epoch": 3.1014420686225757,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044182378202844427,
|
|
"loss": 6.0705,
|
|
"mean_token_accuracy": 0.15912513583898544,
|
|
"num_tokens": 6105735.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.556612300872803,
|
|
"epoch": 3.106414719045251,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044156147769159183,
|
|
"loss": 5.9209,
|
|
"mean_token_accuracy": 0.17123427987098694,
|
|
"num_tokens": 6114132.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.464878082275391,
|
|
"epoch": 3.1113873694679266,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004412986715704591,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.16093026995658874,
|
|
"num_tokens": 6124341.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.39031195640564,
|
|
"epoch": 3.116360019890602,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004410353644567777,
|
|
"loss": 5.8475,
|
|
"mean_token_accuracy": 0.17172767966985703,
|
|
"num_tokens": 6134418.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.454799938201904,
|
|
"epoch": 3.121332670313277,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044077155714378836,
|
|
"loss": 5.9079,
|
|
"mean_token_accuracy": 0.17073309421539307,
|
|
"num_tokens": 6143659.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.455532550811768,
|
|
"epoch": 3.1263053207359524,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004405072504262391,
|
|
"loss": 5.9354,
|
|
"mean_token_accuracy": 0.16392235457897186,
|
|
"num_tokens": 6152706.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.53273229598999,
|
|
"epoch": 3.1312779711586276,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000440242445100382,
|
|
"loss": 5.9868,
|
|
"mean_token_accuracy": 0.15757920145988463,
|
|
"num_tokens": 6161390.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.554285621643066,
|
|
"epoch": 3.136250621581303,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043997714196397143,
|
|
"loss": 5.9367,
|
|
"mean_token_accuracy": 0.16726467311382293,
|
|
"num_tokens": 6170982.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.498422384262085,
|
|
"epoch": 3.141223272003978,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043971134181626173,
|
|
"loss": 5.9861,
|
|
"mean_token_accuracy": 0.16328003108501435,
|
|
"num_tokens": 6179356.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.413940811157227,
|
|
"epoch": 3.1461959224266534,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043944504545800396,
|
|
"loss": 5.9475,
|
|
"mean_token_accuracy": 0.16458411663770675,
|
|
"num_tokens": 6188270.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.482053232192993,
|
|
"epoch": 3.1511685728493286,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004391782536914447,
|
|
"loss": 5.8872,
|
|
"mean_token_accuracy": 0.1671218678355217,
|
|
"num_tokens": 6197624.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.486083745956421,
|
|
"epoch": 3.156141223272004,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043891096732032257,
|
|
"loss": 5.9316,
|
|
"mean_token_accuracy": 0.16723951399326326,
|
|
"num_tokens": 6207174.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.333275604248047,
|
|
"epoch": 3.161113873694679,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043864318714986625,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.1811393305659294,
|
|
"num_tokens": 6217939.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.430752229690552,
|
|
"epoch": 3.1660865241173544,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00043837491398679234,
|
|
"loss": 5.9495,
|
|
"mean_token_accuracy": 0.15952398777008056,
|
|
"num_tokens": 6227857.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.43276014328003,
|
|
"epoch": 3.17105917454003,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004381061486393025,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.15977156311273574,
|
|
"num_tokens": 6236832.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.433659791946411,
|
|
"epoch": 3.1760318249627053,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004378368919170809,
|
|
"loss": 5.924,
|
|
"mean_token_accuracy": 0.16674742102622986,
|
|
"num_tokens": 6245879.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.444940614700317,
|
|
"epoch": 3.1810044753853806,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043756714463129244,
|
|
"loss": 5.9524,
|
|
"mean_token_accuracy": 0.16111271530389787,
|
|
"num_tokens": 6256350.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.322683238983155,
|
|
"epoch": 3.185977125808056,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004372969075945796,
|
|
"loss": 5.7193,
|
|
"mean_token_accuracy": 0.18595454543828965,
|
|
"num_tokens": 6266085.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.514824914932251,
|
|
"epoch": 3.190949776230731,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004370261816210604,
|
|
"loss": 5.9883,
|
|
"mean_token_accuracy": 0.1657641738653183,
|
|
"num_tokens": 6276499.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.483495855331421,
|
|
"epoch": 3.1959224266534063,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004367549675263258,
|
|
"loss": 5.9818,
|
|
"mean_token_accuracy": 0.15906689018011094,
|
|
"num_tokens": 6286630.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.414402818679809,
|
|
"epoch": 3.2008950770760816,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004364832661274373,
|
|
"loss": 5.9431,
|
|
"mean_token_accuracy": 0.1629439949989319,
|
|
"num_tokens": 6295907.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.411798000335693,
|
|
"epoch": 3.205867727498757,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043621107824292444,
|
|
"loss": 5.8737,
|
|
"mean_token_accuracy": 0.1676173597574234,
|
|
"num_tokens": 6304304.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.369206428527832,
|
|
"epoch": 3.210840377921432,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004359384046927824,
|
|
"loss": 5.9718,
|
|
"mean_token_accuracy": 0.1576264202594757,
|
|
"num_tokens": 6315041.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.44608359336853,
|
|
"epoch": 3.2158130283441073,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004356652462984693,
|
|
"loss": 5.8619,
|
|
"mean_token_accuracy": 0.16332081258296965,
|
|
"num_tokens": 6324551.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.404736852645874,
|
|
"epoch": 3.2207856787667826,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00043539160388290416,
|
|
"loss": 5.9055,
|
|
"mean_token_accuracy": 0.17017554342746735,
|
|
"num_tokens": 6335512.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.358619642257691,
|
|
"epoch": 3.225758329189458,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004351174782704641,
|
|
"loss": 5.9669,
|
|
"mean_token_accuracy": 0.1605168789625168,
|
|
"num_tokens": 6345987.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.342964363098145,
|
|
"epoch": 3.230730979612133,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043484287028698176,
|
|
"loss": 5.7585,
|
|
"mean_token_accuracy": 0.18538713902235032,
|
|
"num_tokens": 6357617.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.401046705245972,
|
|
"epoch": 3.2357036300348083,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043456778075974306,
|
|
"loss": 5.8533,
|
|
"mean_token_accuracy": 0.16860382705926896,
|
|
"num_tokens": 6366123.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.365039539337158,
|
|
"epoch": 3.240676280457484,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043429221051748487,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.15816969126462938,
|
|
"num_tokens": 6375776.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.520833158493042,
|
|
"epoch": 3.2456489308801593,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043401616039039174,
|
|
"loss": 6.0195,
|
|
"mean_token_accuracy": 0.1491602309048176,
|
|
"num_tokens": 6386235.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.513797664642334,
|
|
"epoch": 3.2506215813028345,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004337396312100944,
|
|
"loss": 5.9796,
|
|
"mean_token_accuracy": 0.1609613999724388,
|
|
"num_tokens": 6396919.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.435710144042969,
|
|
"epoch": 3.2555942317255098,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004334626238096665,
|
|
"loss": 5.9217,
|
|
"mean_token_accuracy": 0.1696544349193573,
|
|
"num_tokens": 6406769.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.378097057342529,
|
|
"epoch": 3.260566882148185,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004331851390236227,
|
|
"loss": 5.9489,
|
|
"mean_token_accuracy": 0.17008643448352814,
|
|
"num_tokens": 6416208.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.444663047790527,
|
|
"epoch": 3.2655395325708603,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043290717768791533,
|
|
"loss": 5.9156,
|
|
"mean_token_accuracy": 0.16723252832889557,
|
|
"num_tokens": 6426125.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.42572808265686,
|
|
"epoch": 3.2705121829935355,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043262874063993285,
|
|
"loss": 5.8753,
|
|
"mean_token_accuracy": 0.15709270685911178,
|
|
"num_tokens": 6436671.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.374664449691773,
|
|
"epoch": 3.2754848334162108,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004323498287184967,
|
|
"loss": 5.9808,
|
|
"mean_token_accuracy": 0.16401388645172119,
|
|
"num_tokens": 6447396.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.453426933288574,
|
|
"epoch": 3.280457483838886,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043207044276385884,
|
|
"loss": 5.94,
|
|
"mean_token_accuracy": 0.16585720628499984,
|
|
"num_tokens": 6457045.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.393131113052368,
|
|
"epoch": 3.2854301342615613,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043179058361769924,
|
|
"loss": 5.9504,
|
|
"mean_token_accuracy": 0.16813024878501892,
|
|
"num_tokens": 6466741.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.276395845413208,
|
|
"epoch": 3.2904027846842365,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004315102521231237,
|
|
"loss": 5.6878,
|
|
"mean_token_accuracy": 0.19100395441055298,
|
|
"num_tokens": 6477769.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.501129055023194,
|
|
"epoch": 3.295375435106912,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004312294491246609,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.15885487496852874,
|
|
"num_tokens": 6486991.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.408685350418091,
|
|
"epoch": 3.3003480855295875,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004309481754682597,
|
|
"loss": 5.9203,
|
|
"mean_token_accuracy": 0.16346026659011842,
|
|
"num_tokens": 6496772.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.451060962677002,
|
|
"epoch": 3.3053207359522627,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043066643200128724,
|
|
"loss": 5.9517,
|
|
"mean_token_accuracy": 0.15985526889562607,
|
|
"num_tokens": 6507577.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.4284443855285645,
|
|
"epoch": 3.310293386374938,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004303842195725259,
|
|
"loss": 5.985,
|
|
"mean_token_accuracy": 0.16049106568098068,
|
|
"num_tokens": 6518040.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.407729387283325,
|
|
"epoch": 3.3152660367976132,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00043010153903217074,
|
|
"loss": 5.9242,
|
|
"mean_token_accuracy": 0.16319386959075927,
|
|
"num_tokens": 6529261.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.436709070205689,
|
|
"epoch": 3.3202386872202885,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004298183912318273,
|
|
"loss": 5.8368,
|
|
"mean_token_accuracy": 0.17007798701524734,
|
|
"num_tokens": 6537527.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.423614072799682,
|
|
"epoch": 3.3252113376429637,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042953477702450845,
|
|
"loss": 5.9751,
|
|
"mean_token_accuracy": 0.16536534130573272,
|
|
"num_tokens": 6547205.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.393251609802246,
|
|
"epoch": 3.330183988065639,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004292506972646324,
|
|
"loss": 6.0198,
|
|
"mean_token_accuracy": 0.16235229820013047,
|
|
"num_tokens": 6556328.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.457984209060669,
|
|
"epoch": 3.3351566384883142,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00042896615280802006,
|
|
"loss": 5.9845,
|
|
"mean_token_accuracy": 0.1605748437345028,
|
|
"num_tokens": 6566414.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.315146446228027,
|
|
"epoch": 3.3401292889109895,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00042868114451189164,
|
|
"loss": 5.8018,
|
|
"mean_token_accuracy": 0.18632688671350478,
|
|
"num_tokens": 6577365.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.4301759719848635,
|
|
"epoch": 3.3451019393336647,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004283956732348655,
|
|
"loss": 5.9237,
|
|
"mean_token_accuracy": 0.1652536004781723,
|
|
"num_tokens": 6586647.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.392470407485962,
|
|
"epoch": 3.35007458975634,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042810973983695426,
|
|
"loss": 6.0159,
|
|
"mean_token_accuracy": 0.15995876342058182,
|
|
"num_tokens": 6596955.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.448505878448486,
|
|
"epoch": 3.3550472401790152,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004278233451795629,
|
|
"loss": 5.8764,
|
|
"mean_token_accuracy": 0.17022366523742677,
|
|
"num_tokens": 6606161.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.282578086853027,
|
|
"epoch": 3.3600198906016905,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000427536490125486,
|
|
"loss": 5.7794,
|
|
"mean_token_accuracy": 0.1797480121254921,
|
|
"num_tokens": 6616540.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.393112325668335,
|
|
"epoch": 3.364992541024366,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042724917553890507,
|
|
"loss": 5.9166,
|
|
"mean_token_accuracy": 0.1685267597436905,
|
|
"num_tokens": 6626109.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.4035368919372555,
|
|
"epoch": 3.3699651914470414,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042696140228538616,
|
|
"loss": 5.9017,
|
|
"mean_token_accuracy": 0.1717073306441307,
|
|
"num_tokens": 6635576.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.385404062271118,
|
|
"epoch": 3.3749378418697167,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00042667317123187696,
|
|
"loss": 6.008,
|
|
"mean_token_accuracy": 0.16431225687265397,
|
|
"num_tokens": 6645728.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.394898509979248,
|
|
"epoch": 3.379910492292392,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042638448324670424,
|
|
"loss": 5.886,
|
|
"mean_token_accuracy": 0.16979603618383407,
|
|
"num_tokens": 6654538.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.352716732025146,
|
|
"epoch": 3.384883142715067,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004260953391995716,
|
|
"loss": 5.8379,
|
|
"mean_token_accuracy": 0.17085248678922654,
|
|
"num_tokens": 6664811.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.415171194076538,
|
|
"epoch": 3.3898557931377424,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042580573996155646,
|
|
"loss": 5.9571,
|
|
"mean_token_accuracy": 0.17273338437080382,
|
|
"num_tokens": 6674185.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.464240598678589,
|
|
"epoch": 3.3948284435604177,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004255156864051074,
|
|
"loss": 5.9517,
|
|
"mean_token_accuracy": 0.16041996479034423,
|
|
"num_tokens": 6684103.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.211953639984131,
|
|
"epoch": 3.399801093983093,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042522517940404186,
|
|
"loss": 5.6343,
|
|
"mean_token_accuracy": 0.195681132376194,
|
|
"num_tokens": 6693631.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.397243309020996,
|
|
"epoch": 3.404773744405768,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004249342198335433,
|
|
"loss": 5.9679,
|
|
"mean_token_accuracy": 0.16505518108606337,
|
|
"num_tokens": 6704465.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.378099489212036,
|
|
"epoch": 3.4097463948284434,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042464280857015856,
|
|
"loss": 5.9139,
|
|
"mean_token_accuracy": 0.17288842797279358,
|
|
"num_tokens": 6713592.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.392910575866699,
|
|
"epoch": 3.4147190452511187,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042435094649179525,
|
|
"loss": 5.9644,
|
|
"mean_token_accuracy": 0.16168729811906815,
|
|
"num_tokens": 6726252.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.452109241485596,
|
|
"epoch": 3.419691695673794,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004240586344777191,
|
|
"loss": 6.0317,
|
|
"mean_token_accuracy": 0.15827949047088624,
|
|
"num_tokens": 6736301.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.4550292015075685,
|
|
"epoch": 3.4246643460964696,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004237658734085514,
|
|
"loss": 5.989,
|
|
"mean_token_accuracy": 0.1588227167725563,
|
|
"num_tokens": 6745706.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.441744375228882,
|
|
"epoch": 3.429636996519145,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042347266416626624,
|
|
"loss": 5.9844,
|
|
"mean_token_accuracy": 0.15569704174995422,
|
|
"num_tokens": 6755811.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.48943943977356,
|
|
"epoch": 3.43460964694182,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004231790076341877,
|
|
"loss": 5.9451,
|
|
"mean_token_accuracy": 0.16194335222244263,
|
|
"num_tokens": 6766365.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.347968626022339,
|
|
"epoch": 3.4395822973644954,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004228849046969876,
|
|
"loss": 5.8831,
|
|
"mean_token_accuracy": 0.17766775339841842,
|
|
"num_tokens": 6775806.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.426005077362061,
|
|
"epoch": 3.4445549477871706,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004225903562406826,
|
|
"loss": 5.9294,
|
|
"mean_token_accuracy": 0.16613464653491974,
|
|
"num_tokens": 6785996.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.420203351974488,
|
|
"epoch": 3.449527598209846,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004222953631526313,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.17784045040607452,
|
|
"num_tokens": 6795528.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.354560136795044,
|
|
"epoch": 3.454500248632521,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000421999926321532,
|
|
"loss": 5.907,
|
|
"mean_token_accuracy": 0.16833371818065643,
|
|
"num_tokens": 6804592.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.34170446395874,
|
|
"epoch": 3.4594728990551964,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004217040466374199,
|
|
"loss": 5.922,
|
|
"mean_token_accuracy": 0.1649264484643936,
|
|
"num_tokens": 6813242.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.472457551956177,
|
|
"epoch": 3.4644455494778716,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00042140772499166403,
|
|
"loss": 5.9516,
|
|
"mean_token_accuracy": 0.16570166498422623,
|
|
"num_tokens": 6823223.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.257113456726074,
|
|
"epoch": 3.469418199900547,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004211109622769652,
|
|
"loss": 5.7742,
|
|
"mean_token_accuracy": 0.18082345128059388,
|
|
"num_tokens": 6832425.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.326738071441651,
|
|
"epoch": 3.474390850323222,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042081375938735283,
|
|
"loss": 5.9527,
|
|
"mean_token_accuracy": 0.16323843449354172,
|
|
"num_tokens": 6842549.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.451035881042481,
|
|
"epoch": 3.4793635007458974,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042051611721818233,
|
|
"loss": 6.0545,
|
|
"mean_token_accuracy": 0.15942852050065995,
|
|
"num_tokens": 6851652.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.4793635007458974,
|
|
"eval_entropy": 6.3125124362207226,
|
|
"eval_loss": 6.578086853027344,
|
|
"eval_mean_token_accuracy": 0.14198629304011295,
|
|
"eval_num_tokens": 6851652.0,
|
|
"eval_runtime": 1.9416,
|
|
"eval_samples_per_second": 1784.615,
|
|
"eval_steps_per_second": 223.528,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.416307783126831,
|
|
"epoch": 3.4843361511685726,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042021803666613273,
|
|
"loss": 5.9014,
|
|
"mean_token_accuracy": 0.16855714917182923,
|
|
"num_tokens": 6860716.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.314413595199585,
|
|
"epoch": 3.489308801591248,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041991951862920365,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.16993083357810973,
|
|
"num_tokens": 6870074.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.4595824718475345,
|
|
"epoch": 3.4942814520139236,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004196205640067124,
|
|
"loss": 6.1477,
|
|
"mean_token_accuracy": 0.1536518394947052,
|
|
"num_tokens": 6880569.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.5027154922485355,
|
|
"epoch": 3.499254102436599,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000419321173699292,
|
|
"loss": 6.0135,
|
|
"mean_token_accuracy": 0.15570675879716872,
|
|
"num_tokens": 6890522.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.380772399902344,
|
|
"epoch": 3.504226752859274,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004190213486088877,
|
|
"loss": 5.9538,
|
|
"mean_token_accuracy": 0.16360536515712737,
|
|
"num_tokens": 6900388.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.3916100025177,
|
|
"epoch": 3.5091994032819493,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004187210896387546,
|
|
"loss": 5.9144,
|
|
"mean_token_accuracy": 0.1677800089120865,
|
|
"num_tokens": 6910467.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.355901145935059,
|
|
"epoch": 3.5141720537046246,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004184203976934552,
|
|
"loss": 5.9466,
|
|
"mean_token_accuracy": 0.16735238581895828,
|
|
"num_tokens": 6919370.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.301413726806641,
|
|
"epoch": 3.5191447041273,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041811927367885605,
|
|
"loss": 5.7641,
|
|
"mean_token_accuracy": 0.18181305825710298,
|
|
"num_tokens": 6929688.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.406485509872437,
|
|
"epoch": 3.524117354549975,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004178177185021254,
|
|
"loss": 5.9778,
|
|
"mean_token_accuracy": 0.16711320579051972,
|
|
"num_tokens": 6939861.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.403860950469971,
|
|
"epoch": 3.5290900049726504,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004175157330717305,
|
|
"loss": 6.0102,
|
|
"mean_token_accuracy": 0.15020001083612441,
|
|
"num_tokens": 6949746.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.460002040863037,
|
|
"epoch": 3.5340626553953256,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000417213318297435,
|
|
"loss": 5.9391,
|
|
"mean_token_accuracy": 0.166162671148777,
|
|
"num_tokens": 6961408.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.383548593521118,
|
|
"epoch": 3.539035305818001,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041691047509029555,
|
|
"loss": 5.9989,
|
|
"mean_token_accuracy": 0.1676156848669052,
|
|
"num_tokens": 6972003.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.418531703948974,
|
|
"epoch": 3.5440079562406765,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041660720436265987,
|
|
"loss": 5.9265,
|
|
"mean_token_accuracy": 0.16218527853488923,
|
|
"num_tokens": 6983528.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.420976638793945,
|
|
"epoch": 3.548980606663352,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004163035070281634,
|
|
"loss": 5.9639,
|
|
"mean_token_accuracy": 0.16881392747163773,
|
|
"num_tokens": 6992766.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.382440233230591,
|
|
"epoch": 3.553953257086027,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041599938400172686,
|
|
"loss": 5.8506,
|
|
"mean_token_accuracy": 0.17068996876478196,
|
|
"num_tokens": 7001797.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.442717027664185,
|
|
"epoch": 3.5589259075087023,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041569483619955354,
|
|
"loss": 5.8986,
|
|
"mean_token_accuracy": 0.16668633818626405,
|
|
"num_tokens": 7011215.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.3890033721923825,
|
|
"epoch": 3.5638985579313776,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00041538986453912616,
|
|
"loss": 5.9508,
|
|
"mean_token_accuracy": 0.1608192056417465,
|
|
"num_tokens": 7021513.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.333178472518921,
|
|
"epoch": 3.568871208354053,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004150844699392045,
|
|
"loss": 5.95,
|
|
"mean_token_accuracy": 0.16285506039857864,
|
|
"num_tokens": 7031654.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.339065647125244,
|
|
"epoch": 3.573843858776728,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041477865331982264,
|
|
"loss": 5.9206,
|
|
"mean_token_accuracy": 0.16879623383283615,
|
|
"num_tokens": 7040614.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.361328411102295,
|
|
"epoch": 3.5788165091994033,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004144724156022857,
|
|
"loss": 5.9356,
|
|
"mean_token_accuracy": 0.16621116250753404,
|
|
"num_tokens": 7050877.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.527002620697021,
|
|
"epoch": 3.5837891596220786,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041416575770916763,
|
|
"loss": 6.1072,
|
|
"mean_token_accuracy": 0.1521662026643753,
|
|
"num_tokens": 7061330.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.465698766708374,
|
|
"epoch": 3.588761810044754,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00041385868056430827,
|
|
"loss": 5.9821,
|
|
"mean_token_accuracy": 0.16539173275232316,
|
|
"num_tokens": 7071290.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.401636362075806,
|
|
"epoch": 3.593734460467429,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004135511850928102,
|
|
"loss": 5.9798,
|
|
"mean_token_accuracy": 0.1601047024130821,
|
|
"num_tokens": 7080112.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.428067588806153,
|
|
"epoch": 3.5987071108901043,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00041324327222103655,
|
|
"loss": 6.1122,
|
|
"mean_token_accuracy": 0.15216168016195297,
|
|
"num_tokens": 7091467.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.400322723388672,
|
|
"epoch": 3.6036797613127796,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041293494287660785,
|
|
"loss": 5.7521,
|
|
"mean_token_accuracy": 0.18333356976509094,
|
|
"num_tokens": 7101744.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.4175316333770756,
|
|
"epoch": 3.608652411735455,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00041262619798839914,
|
|
"loss": 5.9426,
|
|
"mean_token_accuracy": 0.16271291375160218,
|
|
"num_tokens": 7112854.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.269862174987793,
|
|
"epoch": 3.61362506215813,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041231703848653746,
|
|
"loss": 5.944,
|
|
"mean_token_accuracy": 0.17047446221113205,
|
|
"num_tokens": 7121613.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.3627176761627195,
|
|
"epoch": 3.6185977125808053,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00041200746530239903,
|
|
"loss": 6.0065,
|
|
"mean_token_accuracy": 0.16295944303274154,
|
|
"num_tokens": 7131357.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.327699661254883,
|
|
"epoch": 3.623570363003481,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004116974793686061,
|
|
"loss": 5.8695,
|
|
"mean_token_accuracy": 0.17704739719629287,
|
|
"num_tokens": 7140200.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.372074031829834,
|
|
"epoch": 3.6285430134261563,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004113870816190244,
|
|
"loss": 5.9623,
|
|
"mean_token_accuracy": 0.16677413880825043,
|
|
"num_tokens": 7149367.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.290438604354859,
|
|
"epoch": 3.6335156638488315,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004110762729887605,
|
|
"loss": 5.9361,
|
|
"mean_token_accuracy": 0.16979965567588806,
|
|
"num_tokens": 7158088.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.360591411590576,
|
|
"epoch": 3.6384883142715068,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00041076505441415873,
|
|
"loss": 5.9384,
|
|
"mean_token_accuracy": 0.170064015686512,
|
|
"num_tokens": 7168141.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.458514738082886,
|
|
"epoch": 3.643460964694182,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004104534268327982,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.16960454136133193,
|
|
"num_tokens": 7177646.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.434624910354614,
|
|
"epoch": 3.6484336151168573,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004101413911834904,
|
|
"loss": 6.0201,
|
|
"mean_token_accuracy": 0.15946577936410905,
|
|
"num_tokens": 7186445.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.420453310012817,
|
|
"epoch": 3.6534062655395325,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040982894840627614,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.16720378398895264,
|
|
"num_tokens": 7195812.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.323687505722046,
|
|
"epoch": 3.6583789159622078,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004095160994424228,
|
|
"loss": 5.8125,
|
|
"mean_token_accuracy": 0.1727418228983879,
|
|
"num_tokens": 7204756.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.39320969581604,
|
|
"epoch": 3.663351566384883,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040920284523442125,
|
|
"loss": 6.0039,
|
|
"mean_token_accuracy": 0.16734870672225952,
|
|
"num_tokens": 7214708.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.365419101715088,
|
|
"epoch": 3.6683242168075583,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00040888918672598337,
|
|
"loss": 5.9906,
|
|
"mean_token_accuracy": 0.1619105100631714,
|
|
"num_tokens": 7224623.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.399313688278198,
|
|
"epoch": 3.673296867230234,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000408575124862039,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.16729727536439895,
|
|
"num_tokens": 7235228.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.383471393585205,
|
|
"epoch": 3.678269517652909,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004082606605887331,
|
|
"loss": 6.0039,
|
|
"mean_token_accuracy": 0.17052146941423416,
|
|
"num_tokens": 7245318.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.130453062057495,
|
|
"epoch": 3.6832421680755845,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040794579485342304,
|
|
"loss": 5.5872,
|
|
"mean_token_accuracy": 0.19959189146757125,
|
|
"num_tokens": 7255468.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.409870290756226,
|
|
"epoch": 3.6882148184982597,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00040763052860467547,
|
|
"loss": 5.9798,
|
|
"mean_token_accuracy": 0.16353940814733506,
|
|
"num_tokens": 7265006.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.352113103866577,
|
|
"epoch": 3.693187468920935,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004073148627922637,
|
|
"loss": 5.9379,
|
|
"mean_token_accuracy": 0.16528619974851608,
|
|
"num_tokens": 7273835.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.427479648590088,
|
|
"epoch": 3.69816011934361,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004069987983671648,
|
|
"loss": 6.0431,
|
|
"mean_token_accuracy": 0.15741072297096254,
|
|
"num_tokens": 7284617.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.299086284637451,
|
|
"epoch": 3.7031327697662855,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040668233628155664,
|
|
"loss": 5.7698,
|
|
"mean_token_accuracy": 0.18414730727672576,
|
|
"num_tokens": 7293707.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.404382658004761,
|
|
"epoch": 3.7081054201889607,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004063654774888153,
|
|
"loss": 5.9869,
|
|
"mean_token_accuracy": 0.15650054216384887,
|
|
"num_tokens": 7303227.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.338438606262207,
|
|
"epoch": 3.713078070611636,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004060482229435116,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.17438423931598662,
|
|
"num_tokens": 7312782.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.350795412063599,
|
|
"epoch": 3.7180507210343112,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004057305736014089,
|
|
"loss": 5.9817,
|
|
"mean_token_accuracy": 0.16379624754190444,
|
|
"num_tokens": 7322249.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.431347227096557,
|
|
"epoch": 3.7230233714569865,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004054125304194598,
|
|
"loss": 5.9463,
|
|
"mean_token_accuracy": 0.17110808044672013,
|
|
"num_tokens": 7331076.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.384088516235352,
|
|
"epoch": 3.7279960218796617,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040509409435580355,
|
|
"loss": 5.9133,
|
|
"mean_token_accuracy": 0.17103550285100938,
|
|
"num_tokens": 7340039.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.322669744491577,
|
|
"epoch": 3.732968672302337,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004047752663697627,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.1673117145895958,
|
|
"num_tokens": 7349300.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.486397361755371,
|
|
"epoch": 3.7379413227250122,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004044560474218409,
|
|
"loss": 6.0191,
|
|
"mean_token_accuracy": 0.16359854191541673,
|
|
"num_tokens": 7358363.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.375521087646485,
|
|
"epoch": 3.7429139731476875,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040413643847371923,
|
|
"loss": 5.9268,
|
|
"mean_token_accuracy": 0.1661538228392601,
|
|
"num_tokens": 7369391.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.340249967575073,
|
|
"epoch": 3.7478866235703627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004038164404882539,
|
|
"loss": 5.9836,
|
|
"mean_token_accuracy": 0.1564318910241127,
|
|
"num_tokens": 7378272.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.374636936187744,
|
|
"epoch": 3.7528592739930384,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004034960544294731,
|
|
"loss": 5.8462,
|
|
"mean_token_accuracy": 0.16795043647289276,
|
|
"num_tokens": 7387590.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.367386102676392,
|
|
"epoch": 3.7578319244157137,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040317528126257413,
|
|
"loss": 5.9489,
|
|
"mean_token_accuracy": 0.1587081864476204,
|
|
"num_tokens": 7397733.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.362104225158691,
|
|
"epoch": 3.762804574838389,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004028541219539205,
|
|
"loss": 5.8326,
|
|
"mean_token_accuracy": 0.16909416615962983,
|
|
"num_tokens": 7406748.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.364098453521729,
|
|
"epoch": 3.767777225261064,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004025325774710391,
|
|
"loss": 6.0433,
|
|
"mean_token_accuracy": 0.16152265369892121,
|
|
"num_tokens": 7416419.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.386119413375854,
|
|
"epoch": 3.7727498756837394,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040221064878261684,
|
|
"loss": 5.901,
|
|
"mean_token_accuracy": 0.16821241676807402,
|
|
"num_tokens": 7425257.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.434759664535522,
|
|
"epoch": 3.7777225261064147,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004018883368584985,
|
|
"loss": 5.9981,
|
|
"mean_token_accuracy": 0.16232035160064698,
|
|
"num_tokens": 7436914.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.390273380279541,
|
|
"epoch": 3.78269517652909,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040156564266968336,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.16545664370059968,
|
|
"num_tokens": 7447568.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.374241733551026,
|
|
"epoch": 3.787667826951765,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040124256718832214,
|
|
"loss": 5.9335,
|
|
"mean_token_accuracy": 0.1695076271891594,
|
|
"num_tokens": 7457717.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.425429105758667,
|
|
"epoch": 3.7926404773744404,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004009191113877141,
|
|
"loss": 5.943,
|
|
"mean_token_accuracy": 0.16454798355698586,
|
|
"num_tokens": 7467123.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.348843812942505,
|
|
"epoch": 3.797613127797116,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004005952762423048,
|
|
"loss": 5.8989,
|
|
"mean_token_accuracy": 0.16960630267858506,
|
|
"num_tokens": 7476987.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.40447506904602,
|
|
"epoch": 3.8025857782197914,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004002710627276821,
|
|
"loss": 6.0012,
|
|
"mean_token_accuracy": 0.16775702238082885,
|
|
"num_tokens": 7487024.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.393280124664306,
|
|
"epoch": 3.8075584286424666,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003999464718205739,
|
|
"loss": 6.0042,
|
|
"mean_token_accuracy": 0.16511411666870118,
|
|
"num_tokens": 7496826.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.3515918254852295,
|
|
"epoch": 3.812531079065142,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003996215044988451,
|
|
"loss": 5.9761,
|
|
"mean_token_accuracy": 0.1630323827266693,
|
|
"num_tokens": 7506981.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.297566604614258,
|
|
"epoch": 3.817503729487817,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00039929616174149454,
|
|
"loss": 5.7782,
|
|
"mean_token_accuracy": 0.17964706420898438,
|
|
"num_tokens": 7516925.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.3655173778533936,
|
|
"epoch": 3.8224763799104924,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039897044452865207,
|
|
"loss": 5.911,
|
|
"mean_token_accuracy": 0.17521550506353378,
|
|
"num_tokens": 7525789.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.301772499084473,
|
|
"epoch": 3.8274490303331676,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003986443538415756,
|
|
"loss": 5.8616,
|
|
"mean_token_accuracy": 0.17220608294010162,
|
|
"num_tokens": 7535907.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.362651395797729,
|
|
"epoch": 3.832421680755843,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003983178906626484,
|
|
"loss": 5.9009,
|
|
"mean_token_accuracy": 0.1723572239279747,
|
|
"num_tokens": 7546985.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.26556625366211,
|
|
"epoch": 3.837394331178518,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039799105597537564,
|
|
"loss": 5.7398,
|
|
"mean_token_accuracy": 0.1846863031387329,
|
|
"num_tokens": 7556912.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.368065547943115,
|
|
"epoch": 3.8423669816011934,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003976638507643818,
|
|
"loss": 5.9553,
|
|
"mean_token_accuracy": 0.16706154942512513,
|
|
"num_tokens": 7567156.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.380944299697876,
|
|
"epoch": 3.8473396320238686,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003973362760154076,
|
|
"loss": 5.9111,
|
|
"mean_token_accuracy": 0.1730001300573349,
|
|
"num_tokens": 7576423.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.410670614242553,
|
|
"epoch": 3.852312282446544,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003970083327153071,
|
|
"loss": 5.8999,
|
|
"mean_token_accuracy": 0.17044544965028763,
|
|
"num_tokens": 7585612.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.247180986404419,
|
|
"epoch": 3.857284932869219,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003966800218520445,
|
|
"loss": 5.9109,
|
|
"mean_token_accuracy": 0.16674140840768814,
|
|
"num_tokens": 7595581.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.397925090789795,
|
|
"epoch": 3.8622575832918944,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003963513444146914,
|
|
"loss": 5.9291,
|
|
"mean_token_accuracy": 0.16625275164842607,
|
|
"num_tokens": 7605202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.408295488357544,
|
|
"epoch": 3.8672302337145696,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003960223013934239,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.17699990868568422,
|
|
"num_tokens": 7615783.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.404464292526245,
|
|
"epoch": 3.872202884137245,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003956928937795192,
|
|
"loss": 6.0316,
|
|
"mean_token_accuracy": 0.15575970709323883,
|
|
"num_tokens": 7626609.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.361887788772583,
|
|
"epoch": 3.8771755345599206,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000395363122565353,
|
|
"loss": 5.9205,
|
|
"mean_token_accuracy": 0.17140279114246368,
|
|
"num_tokens": 7636676.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.393074798583984,
|
|
"epoch": 3.882148184982596,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039503298874439655,
|
|
"loss": 5.9116,
|
|
"mean_token_accuracy": 0.17187585383653642,
|
|
"num_tokens": 7646523.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.287360095977784,
|
|
"epoch": 3.887120835405271,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003947024933112131,
|
|
"loss": 5.8149,
|
|
"mean_token_accuracy": 0.1787031203508377,
|
|
"num_tokens": 7656267.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.367463064193726,
|
|
"epoch": 3.8920934858279463,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003943716372614556,
|
|
"loss": 5.8887,
|
|
"mean_token_accuracy": 0.173692224919796,
|
|
"num_tokens": 7665529.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.413134145736694,
|
|
"epoch": 3.8970661362506216,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003940404215918634,
|
|
"loss": 5.9546,
|
|
"mean_token_accuracy": 0.16816517412662507,
|
|
"num_tokens": 7675477.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.1843101978302,
|
|
"epoch": 3.902038786673297,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039370884730025897,
|
|
"loss": 5.5876,
|
|
"mean_token_accuracy": 0.19785896241664885,
|
|
"num_tokens": 7685675.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.294821262359619,
|
|
"epoch": 3.907011437095972,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003933769153855455,
|
|
"loss": 5.9355,
|
|
"mean_token_accuracy": 0.17053910940885544,
|
|
"num_tokens": 7695821.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.342487907409668,
|
|
"epoch": 3.9119840875186473,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003930446268477034,
|
|
"loss": 5.9269,
|
|
"mean_token_accuracy": 0.16830311566591263,
|
|
"num_tokens": 7706163.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.356036424636841,
|
|
"epoch": 3.9169567379413226,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003927119826877875,
|
|
"loss": 5.8642,
|
|
"mean_token_accuracy": 0.17302054315805435,
|
|
"num_tokens": 7715125.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.331394863128662,
|
|
"epoch": 3.921929388363998,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003923789839079238,
|
|
"loss": 5.8649,
|
|
"mean_token_accuracy": 0.17714743614196776,
|
|
"num_tokens": 7724546.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.214202833175659,
|
|
"epoch": 3.9269020387866735,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00039204563151130683,
|
|
"loss": 5.7117,
|
|
"mean_token_accuracy": 0.18506779670715331,
|
|
"num_tokens": 7734829.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.320785903930664,
|
|
"epoch": 3.931874689209349,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003917119265021965,
|
|
"loss": 5.8858,
|
|
"mean_token_accuracy": 0.1683538407087326,
|
|
"num_tokens": 7744612.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.390184164047241,
|
|
"epoch": 3.936847339632024,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003913778698859146,
|
|
"loss": 5.9691,
|
|
"mean_token_accuracy": 0.16978412866592407,
|
|
"num_tokens": 7754042.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.3747728824615475,
|
|
"epoch": 3.9418199900546993,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003910434626688427,
|
|
"loss": 5.8655,
|
|
"mean_token_accuracy": 0.17298921942710876,
|
|
"num_tokens": 7762608.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.2740521907806395,
|
|
"epoch": 3.9467926404773745,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003907087058584183,
|
|
"loss": 5.8255,
|
|
"mean_token_accuracy": 0.17217137515544892,
|
|
"num_tokens": 7771477.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.292557573318481,
|
|
"epoch": 3.95176529090005,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039037360046313186,
|
|
"loss": 5.8866,
|
|
"mean_token_accuracy": 0.17796075567603112,
|
|
"num_tokens": 7782785.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.504504680633545,
|
|
"epoch": 3.956737941322725,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00039003814749252454,
|
|
"loss": 6.0657,
|
|
"mean_token_accuracy": 0.15771323665976525,
|
|
"num_tokens": 7793572.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.310393619537353,
|
|
"epoch": 3.9617105917454003,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038970234795718424,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.17050646990537643,
|
|
"num_tokens": 7803170.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.337598609924316,
|
|
"epoch": 3.9666832421680756,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000389366202868743,
|
|
"loss": 6.0051,
|
|
"mean_token_accuracy": 0.16291176080703734,
|
|
"num_tokens": 7813445.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.403491687774658,
|
|
"epoch": 3.971655892590751,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003890297132398739,
|
|
"loss": 5.9951,
|
|
"mean_token_accuracy": 0.16686530262231827,
|
|
"num_tokens": 7823089.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.40996036529541,
|
|
"epoch": 3.976628543013426,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003886928800842879,
|
|
"loss": 5.8401,
|
|
"mean_token_accuracy": 0.16771667152643205,
|
|
"num_tokens": 7832487.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.976628543013426,
|
|
"eval_entropy": 6.17021008680493,
|
|
"eval_loss": 6.515535831451416,
|
|
"eval_mean_token_accuracy": 0.1436073448388807,
|
|
"eval_num_tokens": 7832487.0,
|
|
"eval_runtime": 1.9349,
|
|
"eval_samples_per_second": 1790.767,
|
|
"eval_steps_per_second": 224.298,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 6.258194160461426,
|
|
"epoch": 3.9816011934361013,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00038835570441673095,
|
|
"loss": 5.889,
|
|
"mean_token_accuracy": 0.1703777715563774,
|
|
"num_tokens": 7842540.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 6.380897283554077,
|
|
"epoch": 3.9865738438587766,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003880181872529809,
|
|
"loss": 5.9615,
|
|
"mean_token_accuracy": 0.16862134039402008,
|
|
"num_tokens": 7852384.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 6.371260356903076,
|
|
"epoch": 3.991546494281452,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003876803296098443,
|
|
"loss": 5.962,
|
|
"mean_token_accuracy": 0.17095237523317336,
|
|
"num_tokens": 7864082.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.363522958755493,
|
|
"epoch": 3.996519144704127,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003873421325051535,
|
|
"loss": 5.8762,
|
|
"mean_token_accuracy": 0.16869421526789666,
|
|
"num_tokens": 7873677.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 6.398655732472737,
|
|
"epoch": 4.000994530084535,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038700359695776346,
|
|
"loss": 5.8822,
|
|
"mean_token_accuracy": 0.16491092575920951,
|
|
"num_tokens": 7882368.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.41061840057373,
|
|
"epoch": 4.00596718050721,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003866647239875487,
|
|
"loss": 5.5886,
|
|
"mean_token_accuracy": 0.18069569319486617,
|
|
"num_tokens": 7892327.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 6.283297920227051,
|
|
"epoch": 4.010939830929885,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00038632551461540023,
|
|
"loss": 5.5716,
|
|
"mean_token_accuracy": 0.1861691117286682,
|
|
"num_tokens": 7902048.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 6.335134840011596,
|
|
"epoch": 4.015912481352561,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003859859698632228,
|
|
"loss": 5.6264,
|
|
"mean_token_accuracy": 0.18411600291728974,
|
|
"num_tokens": 7911512.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.3484508991241455,
|
|
"epoch": 4.020885131775236,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003856460907539312,
|
|
"loss": 5.7128,
|
|
"mean_token_accuracy": 0.17618379592895508,
|
|
"num_tokens": 7921832.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.348491096496582,
|
|
"epoch": 4.025857782197911,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003853058783114478,
|
|
"loss": 5.6325,
|
|
"mean_token_accuracy": 0.18649775683879852,
|
|
"num_tokens": 7931143.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 6.2124982357025145,
|
|
"epoch": 4.030830432620586,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003849653335606987,
|
|
"loss": 5.5564,
|
|
"mean_token_accuracy": 0.19254778921604157,
|
|
"num_tokens": 7941108.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 6.245998477935791,
|
|
"epoch": 4.035803083043262,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003846244575276117,
|
|
"loss": 5.5929,
|
|
"mean_token_accuracy": 0.18013137876987456,
|
|
"num_tokens": 7951464.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.37117109298706,
|
|
"epoch": 4.040775733465938,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038428325123911205,
|
|
"loss": 5.7102,
|
|
"mean_token_accuracy": 0.17535144090652466,
|
|
"num_tokens": 7961626.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.3305902004241945,
|
|
"epoch": 4.045748383888613,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003839417157231205,
|
|
"loss": 5.6892,
|
|
"mean_token_accuracy": 0.1783040300011635,
|
|
"num_tokens": 7970135.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 6.258138608932495,
|
|
"epoch": 4.050721034311288,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003835998520085492,
|
|
"loss": 5.6509,
|
|
"mean_token_accuracy": 0.1858382374048233,
|
|
"num_tokens": 7982177.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 6.293023204803466,
|
|
"epoch": 4.0556936847339635,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003832576611252993,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.1833986982703209,
|
|
"num_tokens": 7991845.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 6.264206838607788,
|
|
"epoch": 4.060666335156639,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003829151441042572,
|
|
"loss": 5.571,
|
|
"mean_token_accuracy": 0.19048189669847487,
|
|
"num_tokens": 8000731.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 6.308241081237793,
|
|
"epoch": 4.065638985579314,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00038257230197729237,
|
|
"loss": 5.5909,
|
|
"mean_token_accuracy": 0.18780712336301802,
|
|
"num_tokens": 8011137.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 6.323659372329712,
|
|
"epoch": 4.070611636001989,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038222913577725325,
|
|
"loss": 5.6418,
|
|
"mean_token_accuracy": 0.178889462351799,
|
|
"num_tokens": 8020413.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 6.277144813537598,
|
|
"epoch": 4.0755842864246645,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038188564653796487,
|
|
"loss": 5.5745,
|
|
"mean_token_accuracy": 0.1936006113886833,
|
|
"num_tokens": 8029727.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 6.296385383605957,
|
|
"epoch": 4.08055693684734,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003815418352942252,
|
|
"loss": 5.6369,
|
|
"mean_token_accuracy": 0.1840255469083786,
|
|
"num_tokens": 8038707.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 6.170658826828003,
|
|
"epoch": 4.085529587270015,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003811977030818026,
|
|
"loss": 5.4111,
|
|
"mean_token_accuracy": 0.2064014568924904,
|
|
"num_tokens": 8048604.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 6.262324094772339,
|
|
"epoch": 4.09050223769269,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000380853250937432,
|
|
"loss": 5.6107,
|
|
"mean_token_accuracy": 0.19120270907878875,
|
|
"num_tokens": 8058089.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 6.380268001556397,
|
|
"epoch": 4.0954748881153655,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038050847989881236,
|
|
"loss": 5.6308,
|
|
"mean_token_accuracy": 0.18061528354883194,
|
|
"num_tokens": 8068078.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 6.2058351039886475,
|
|
"epoch": 4.100447538538041,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003801633910046035,
|
|
"loss": 5.5081,
|
|
"mean_token_accuracy": 0.1956583395600319,
|
|
"num_tokens": 8076918.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 6.115322494506836,
|
|
"epoch": 4.105420188960716,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003798179852944225,
|
|
"loss": 5.4209,
|
|
"mean_token_accuracy": 0.20926634222269058,
|
|
"num_tokens": 8087104.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 6.247693347930908,
|
|
"epoch": 4.110392839383391,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037947226380884115,
|
|
"loss": 5.699,
|
|
"mean_token_accuracy": 0.17430607825517655,
|
|
"num_tokens": 8096976.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 6.292794418334961,
|
|
"epoch": 4.1153654898060665,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00037912622758938243,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.19050992280244827,
|
|
"num_tokens": 8106707.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 6.314745283126831,
|
|
"epoch": 4.120338140228742,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003787798776785174,
|
|
"loss": 5.6153,
|
|
"mean_token_accuracy": 0.17749105989933014,
|
|
"num_tokens": 8116822.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 6.174899673461914,
|
|
"epoch": 4.125310790651417,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003784332151196622,
|
|
"loss": 5.5531,
|
|
"mean_token_accuracy": 0.190416519343853,
|
|
"num_tokens": 8125677.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 6.129270267486572,
|
|
"epoch": 4.130283441074092,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037808624095717503,
|
|
"loss": 5.4797,
|
|
"mean_token_accuracy": 0.2037482187151909,
|
|
"num_tokens": 8134878.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 6.280631494522095,
|
|
"epoch": 4.1352560914967675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003777389562363526,
|
|
"loss": 5.6198,
|
|
"mean_token_accuracy": 0.18876935988664628,
|
|
"num_tokens": 8144491.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 6.303162145614624,
|
|
"epoch": 4.140228741919443,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037739136200342743,
|
|
"loss": 5.6539,
|
|
"mean_token_accuracy": 0.17601989060640336,
|
|
"num_tokens": 8155163.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 6.234076738357544,
|
|
"epoch": 4.145201392342118,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037704345930556426,
|
|
"loss": 5.6042,
|
|
"mean_token_accuracy": 0.18386063277721404,
|
|
"num_tokens": 8164626.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.394153642654419,
|
|
"epoch": 4.150174042764793,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037669524919085716,
|
|
"loss": 5.7534,
|
|
"mean_token_accuracy": 0.17479306757450103,
|
|
"num_tokens": 8174681.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.407816171646118,
|
|
"epoch": 4.1551466931874685,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003763467327083264,
|
|
"loss": 5.8143,
|
|
"mean_token_accuracy": 0.17562297433614732,
|
|
"num_tokens": 8184627.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 6.182187700271607,
|
|
"epoch": 4.160119343610144,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037599791090791527,
|
|
"loss": 5.4546,
|
|
"mean_token_accuracy": 0.19785990417003632,
|
|
"num_tokens": 8195029.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 6.224386215209961,
|
|
"epoch": 4.165091994032819,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.0003756487848404868,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.19212062060832977,
|
|
"num_tokens": 8204897.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 6.304575061798095,
|
|
"epoch": 4.170064644455495,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003752993555578205,
|
|
"loss": 5.7439,
|
|
"mean_token_accuracy": 0.17398486137390137,
|
|
"num_tokens": 8215859.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 6.297217226028442,
|
|
"epoch": 4.17503729487817,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003749496241126095,
|
|
"loss": 5.5443,
|
|
"mean_token_accuracy": 0.1854538008570671,
|
|
"num_tokens": 8225377.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 6.2563282489776615,
|
|
"epoch": 4.180009945300846,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037459959155845726,
|
|
"loss": 5.718,
|
|
"mean_token_accuracy": 0.1740410566329956,
|
|
"num_tokens": 8235436.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 6.264992141723633,
|
|
"epoch": 4.184982595723521,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003742492589498744,
|
|
"loss": 5.6452,
|
|
"mean_token_accuracy": 0.18348868638277055,
|
|
"num_tokens": 8245129.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 6.2469813346862795,
|
|
"epoch": 4.189955246146196,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037389862734227525,
|
|
"loss": 5.605,
|
|
"mean_token_accuracy": 0.184417462348938,
|
|
"num_tokens": 8254870.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 6.27945728302002,
|
|
"epoch": 4.194927896568871,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037354769779197514,
|
|
"loss": 5.668,
|
|
"mean_token_accuracy": 0.1788850337266922,
|
|
"num_tokens": 8266078.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.288291883468628,
|
|
"epoch": 4.199900546991547,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037319647135618695,
|
|
"loss": 5.7025,
|
|
"mean_token_accuracy": 0.1808929294347763,
|
|
"num_tokens": 8275833.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 6.201287031173706,
|
|
"epoch": 4.204873197414222,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00037284494909301793,
|
|
"loss": 5.6324,
|
|
"mean_token_accuracy": 0.1790034741163254,
|
|
"num_tokens": 8286866.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 6.35926513671875,
|
|
"epoch": 4.209845847836897,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003724931320614663,
|
|
"loss": 5.6453,
|
|
"mean_token_accuracy": 0.18138363659381868,
|
|
"num_tokens": 8296316.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 6.346386337280274,
|
|
"epoch": 4.214818498259572,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037214102132141874,
|
|
"loss": 5.8073,
|
|
"mean_token_accuracy": 0.1693449854850769,
|
|
"num_tokens": 8306023.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 6.16111421585083,
|
|
"epoch": 4.219791148682248,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003717886179336465,
|
|
"loss": 5.5515,
|
|
"mean_token_accuracy": 0.18787566423416138,
|
|
"num_tokens": 8315808.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 6.296284484863281,
|
|
"epoch": 4.224763799104923,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037143592295980243,
|
|
"loss": 5.6252,
|
|
"mean_token_accuracy": 0.18620599210262298,
|
|
"num_tokens": 8324951.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 6.2607152462005615,
|
|
"epoch": 4.229736449527598,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003710829374624181,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.18611691743135453,
|
|
"num_tokens": 8335019.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 6.287782192230225,
|
|
"epoch": 4.2347090999502734,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003707296625048999,
|
|
"loss": 5.6108,
|
|
"mean_token_accuracy": 0.18175425827503205,
|
|
"num_tokens": 8344496.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 6.263432359695434,
|
|
"epoch": 4.239681750372949,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003703760991515264,
|
|
"loss": 5.6375,
|
|
"mean_token_accuracy": 0.17820941656827927,
|
|
"num_tokens": 8353766.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 6.3134359359741214,
|
|
"epoch": 4.244654400795624,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037002224846744526,
|
|
"loss": 5.6249,
|
|
"mean_token_accuracy": 0.17810137569904327,
|
|
"num_tokens": 8363576.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.421724319458008,
|
|
"epoch": 4.249627051218299,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00036966811151866943,
|
|
"loss": 5.7266,
|
|
"mean_token_accuracy": 0.16998927146196366,
|
|
"num_tokens": 8372714.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 6.204234409332275,
|
|
"epoch": 4.2545997016409745,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00036931368937207436,
|
|
"loss": 5.6131,
|
|
"mean_token_accuracy": 0.19146547913551332,
|
|
"num_tokens": 8383563.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 6.169955110549926,
|
|
"epoch": 4.25957235206365,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003689589830953948,
|
|
"loss": 5.5413,
|
|
"mean_token_accuracy": 0.19212117046117783,
|
|
"num_tokens": 8393352.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 6.256896495819092,
|
|
"epoch": 4.264545002486325,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003686039937572213,
|
|
"loss": 5.6783,
|
|
"mean_token_accuracy": 0.1813760906457901,
|
|
"num_tokens": 8403219.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 6.226994037628174,
|
|
"epoch": 4.269517652909,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003682487224269972,
|
|
"loss": 5.5441,
|
|
"mean_token_accuracy": 0.19080247730016708,
|
|
"num_tokens": 8411596.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 6.230297613143921,
|
|
"epoch": 4.2744903033316755,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003678931701750155,
|
|
"loss": 5.6371,
|
|
"mean_token_accuracy": 0.1851443439722061,
|
|
"num_tokens": 8421472.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 6.262417221069336,
|
|
"epoch": 4.279462953754351,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00036753733807241536,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.19577519297599794,
|
|
"num_tokens": 8431489.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 6.0771155834198,
|
|
"epoch": 4.284435604177026,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036718122719117907,
|
|
"loss": 5.4054,
|
|
"mean_token_accuracy": 0.20775347352027893,
|
|
"num_tokens": 8441108.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 6.13216199874878,
|
|
"epoch": 4.289408254599701,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003668248386041288,
|
|
"loss": 5.4634,
|
|
"mean_token_accuracy": 0.20200071036815642,
|
|
"num_tokens": 8451664.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 6.281685447692871,
|
|
"epoch": 4.294380905022377,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00036646817338492334,
|
|
"loss": 5.6578,
|
|
"mean_token_accuracy": 0.18399294912815095,
|
|
"num_tokens": 8460531.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 6.279378128051758,
|
|
"epoch": 4.299353555445053,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003661112326080548,
|
|
"loss": 5.6177,
|
|
"mean_token_accuracy": 0.18892679661512374,
|
|
"num_tokens": 8470018.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 6.159647130966187,
|
|
"epoch": 4.304326205867728,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003657540173488455,
|
|
"loss": 5.569,
|
|
"mean_token_accuracy": 0.192726668715477,
|
|
"num_tokens": 8480095.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 6.283668041229248,
|
|
"epoch": 4.309298856290403,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003653965286834446,
|
|
"loss": 5.7238,
|
|
"mean_token_accuracy": 0.17361934930086137,
|
|
"num_tokens": 8489817.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 6.399185276031494,
|
|
"epoch": 4.314271506713078,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003650387676888251,
|
|
"loss": 5.703,
|
|
"mean_token_accuracy": 0.17805413156747818,
|
|
"num_tokens": 8499333.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 6.267030525207519,
|
|
"epoch": 4.319244157135754,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003646807354427802,
|
|
"loss": 5.6687,
|
|
"mean_token_accuracy": 0.17189670503139495,
|
|
"num_tokens": 8508732.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 6.230120277404785,
|
|
"epoch": 4.324216807558429,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00036432243302392025,
|
|
"loss": 5.6849,
|
|
"mean_token_accuracy": 0.17952397912740709,
|
|
"num_tokens": 8519525.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 6.199698209762573,
|
|
"epoch": 4.329189457981104,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036396386151166975,
|
|
"loss": 5.548,
|
|
"mean_token_accuracy": 0.19641838371753692,
|
|
"num_tokens": 8529559.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 6.292362022399902,
|
|
"epoch": 4.334162108403779,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003636050219862638,
|
|
"loss": 5.5882,
|
|
"mean_token_accuracy": 0.17900745421648026,
|
|
"num_tokens": 8538615.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 6.22099027633667,
|
|
"epoch": 4.339134758826455,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00036324591552874493,
|
|
"loss": 5.5837,
|
|
"mean_token_accuracy": 0.18826409429311752,
|
|
"num_tokens": 8548432.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 6.158109092712403,
|
|
"epoch": 4.34410740924913,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00036288654322095973,
|
|
"loss": 5.5399,
|
|
"mean_token_accuracy": 0.19297979921102523,
|
|
"num_tokens": 8558216.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 6.24582691192627,
|
|
"epoch": 4.349080059671805,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036252690614555557,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.19248410761356355,
|
|
"num_tokens": 8568234.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 6.256908750534057,
|
|
"epoch": 4.35405271009448,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003621670053859778,
|
|
"loss": 5.6569,
|
|
"mean_token_accuracy": 0.17909654676914216,
|
|
"num_tokens": 8578140.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 6.270287036895752,
|
|
"epoch": 4.359025360517156,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000361806842026466,
|
|
"loss": 5.6619,
|
|
"mean_token_accuracy": 0.17892567664384842,
|
|
"num_tokens": 8587976.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 6.255285358428955,
|
|
"epoch": 4.363998010939831,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036144641715205077,
|
|
"loss": 5.6661,
|
|
"mean_token_accuracy": 0.18682259917259217,
|
|
"num_tokens": 8597739.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 6.2586853981018065,
|
|
"epoch": 4.368970661362506,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003610857318485507,
|
|
"loss": 5.7131,
|
|
"mean_token_accuracy": 0.17843175679445267,
|
|
"num_tokens": 8608609.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 6.218613243103027,
|
|
"epoch": 4.373943311785181,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036072478720256885,
|
|
"loss": 5.6443,
|
|
"mean_token_accuracy": 0.180078162252903,
|
|
"num_tokens": 8617870.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 6.266687250137329,
|
|
"epoch": 4.378915962207857,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003603635843014895,
|
|
"loss": 5.641,
|
|
"mean_token_accuracy": 0.18192813396453858,
|
|
"num_tokens": 8627811.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 6.299219417572021,
|
|
"epoch": 4.383888612630532,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036000212423347523,
|
|
"loss": 5.6946,
|
|
"mean_token_accuracy": 0.18235698342323303,
|
|
"num_tokens": 8636989.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 6.257914209365845,
|
|
"epoch": 4.388861263053207,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000359640408087463,
|
|
"loss": 5.5587,
|
|
"mean_token_accuracy": 0.1904021456837654,
|
|
"num_tokens": 8646090.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 6.233590698242187,
|
|
"epoch": 4.393833913475882,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003592784369531615,
|
|
"loss": 5.543,
|
|
"mean_token_accuracy": 0.19233611673116685,
|
|
"num_tokens": 8656244.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 6.142570161819458,
|
|
"epoch": 4.398806563898558,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035891621192104764,
|
|
"loss": 5.5292,
|
|
"mean_token_accuracy": 0.1926055371761322,
|
|
"num_tokens": 8665783.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 6.27254581451416,
|
|
"epoch": 4.403779214321233,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00035855373408236294,
|
|
"loss": 5.716,
|
|
"mean_token_accuracy": 0.17484548091888427,
|
|
"num_tokens": 8676736.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 6.279010486602783,
|
|
"epoch": 4.408751864743908,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003581910045291107,
|
|
"loss": 5.6341,
|
|
"mean_token_accuracy": 0.18245839327573776,
|
|
"num_tokens": 8687035.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 6.304026889801025,
|
|
"epoch": 4.413724515166583,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003578280243540526,
|
|
"loss": 5.7012,
|
|
"mean_token_accuracy": 0.17565103471279145,
|
|
"num_tokens": 8696939.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 6.201478433609009,
|
|
"epoch": 4.4186971655892595,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00035746479465070515,
|
|
"loss": 5.6811,
|
|
"mean_token_accuracy": 0.18189298808574678,
|
|
"num_tokens": 8707213.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 6.235442733764648,
|
|
"epoch": 4.423669816011935,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00035710131651333683,
|
|
"loss": 5.6287,
|
|
"mean_token_accuracy": 0.18110422939062118,
|
|
"num_tokens": 8717066.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 6.317096185684204,
|
|
"epoch": 4.42864246643461,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035673759103696447,
|
|
"loss": 5.6568,
|
|
"mean_token_accuracy": 0.17657845318317414,
|
|
"num_tokens": 8727421.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 6.29308590888977,
|
|
"epoch": 4.433615116857285,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003563736193173499,
|
|
"loss": 5.6864,
|
|
"mean_token_accuracy": 0.18429041057825088,
|
|
"num_tokens": 8736086.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 6.308242750167847,
|
|
"epoch": 4.4385877672799605,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003560094024509969,
|
|
"loss": 5.7259,
|
|
"mean_token_accuracy": 0.18213173151016235,
|
|
"num_tokens": 8746496.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 6.221246337890625,
|
|
"epoch": 4.443560417702636,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035564494153514777,
|
|
"loss": 5.5783,
|
|
"mean_token_accuracy": 0.19261206835508346,
|
|
"num_tokens": 8755528.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 6.310869407653809,
|
|
"epoch": 4.448533068125311,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035528023766778,
|
|
"loss": 5.6408,
|
|
"mean_token_accuracy": 0.17598621994256974,
|
|
"num_tokens": 8765486.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 6.298426580429077,
|
|
"epoch": 4.453505718547986,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003549152919476031,
|
|
"loss": 5.7142,
|
|
"mean_token_accuracy": 0.1808569997549057,
|
|
"num_tokens": 8775315.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 6.2502601146698,
|
|
"epoch": 4.4584783689706615,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035455010547405513,
|
|
"loss": 5.683,
|
|
"mean_token_accuracy": 0.17543030232191087,
|
|
"num_tokens": 8784976.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 6.26897029876709,
|
|
"epoch": 4.463451019393337,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035418467934729934,
|
|
"loss": 5.73,
|
|
"mean_token_accuracy": 0.1777380734682083,
|
|
"num_tokens": 8795412.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 6.224596166610718,
|
|
"epoch": 4.468423669816012,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035381901466822113,
|
|
"loss": 5.6082,
|
|
"mean_token_accuracy": 0.18462185710668563,
|
|
"num_tokens": 8805277.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 6.258960390090943,
|
|
"epoch": 4.473396320238687,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003534531125384244,
|
|
"loss": 5.625,
|
|
"mean_token_accuracy": 0.1818544238805771,
|
|
"num_tokens": 8814308.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.473396320238687,
|
|
"eval_entropy": 6.073968022649739,
|
|
"eval_loss": 6.506396770477295,
|
|
"eval_mean_token_accuracy": 0.14441241653558845,
|
|
"eval_num_tokens": 8814308.0,
|
|
"eval_runtime": 1.9332,
|
|
"eval_samples_per_second": 1792.366,
|
|
"eval_steps_per_second": 224.498,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 6.213192749023437,
|
|
"epoch": 4.4783689706613625,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00035308697406022865,
|
|
"loss": 5.6973,
|
|
"mean_token_accuracy": 0.18164609968662263,
|
|
"num_tokens": 8824836.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 6.194040107727051,
|
|
"epoch": 4.483341621084038,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00035272060033666503,
|
|
"loss": 5.5355,
|
|
"mean_token_accuracy": 0.18553720861673356,
|
|
"num_tokens": 8834390.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 6.293767261505127,
|
|
"epoch": 4.488314271506713,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003523539924714738,
|
|
"loss": 5.7374,
|
|
"mean_token_accuracy": 0.17858339697122574,
|
|
"num_tokens": 8844266.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 6.284969758987427,
|
|
"epoch": 4.493286921929388,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035198715156910016,
|
|
"loss": 5.6275,
|
|
"mean_token_accuracy": 0.18307457268238067,
|
|
"num_tokens": 8852876.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 6.09525876045227,
|
|
"epoch": 4.4982595723520635,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00035162007873469175,
|
|
"loss": 5.3712,
|
|
"mean_token_accuracy": 0.2060387387871742,
|
|
"num_tokens": 8862861.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 6.252058362960815,
|
|
"epoch": 4.503232222774739,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035125277507409483,
|
|
"loss": 5.5953,
|
|
"mean_token_accuracy": 0.18671054393053055,
|
|
"num_tokens": 8872177.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 6.153684377670288,
|
|
"epoch": 4.508204873197414,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00035088524169385075,
|
|
"loss": 5.6283,
|
|
"mean_token_accuracy": 0.1844818711280823,
|
|
"num_tokens": 8882202.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 6.209964084625244,
|
|
"epoch": 4.513177523620089,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035051747970119354,
|
|
"loss": 5.6353,
|
|
"mean_token_accuracy": 0.1844743862748146,
|
|
"num_tokens": 8892361.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 6.288511753082275,
|
|
"epoch": 4.5181501740427645,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00035014949020404523,
|
|
"loss": 5.7435,
|
|
"mean_token_accuracy": 0.17302019149065018,
|
|
"num_tokens": 8903014.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 6.263081502914429,
|
|
"epoch": 4.52312282446544,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00034978127431101365,
|
|
"loss": 5.7141,
|
|
"mean_token_accuracy": 0.17469271570444106,
|
|
"num_tokens": 8912145.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 6.249042606353759,
|
|
"epoch": 4.528095474888115,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00034941283313138865,
|
|
"loss": 5.6232,
|
|
"mean_token_accuracy": 0.18559577018022538,
|
|
"num_tokens": 8921748.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 6.244568490982056,
|
|
"epoch": 4.53306812531079,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003490441677751387,
|
|
"loss": 5.7046,
|
|
"mean_token_accuracy": 0.18024387806653977,
|
|
"num_tokens": 8932327.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 6.24149866104126,
|
|
"epoch": 4.5380407757334655,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034867527935290773,
|
|
"loss": 5.711,
|
|
"mean_token_accuracy": 0.17897280007600785,
|
|
"num_tokens": 8942781.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 6.32926459312439,
|
|
"epoch": 4.543013426156142,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034830616897601146,
|
|
"loss": 5.6792,
|
|
"mean_token_accuracy": 0.17815803587436677,
|
|
"num_tokens": 8951887.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 6.258956050872802,
|
|
"epoch": 4.547986076578816,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003479368377564344,
|
|
"loss": 5.7279,
|
|
"mean_token_accuracy": 0.1773396611213684,
|
|
"num_tokens": 8961454.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 6.201504516601562,
|
|
"epoch": 4.552958727001492,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003475672868068265,
|
|
"loss": 5.5916,
|
|
"mean_token_accuracy": 0.18211541175842286,
|
|
"num_tokens": 8970218.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 6.28542423248291,
|
|
"epoch": 4.557931377424167,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034719751724049955,
|
|
"loss": 5.6332,
|
|
"mean_token_accuracy": 0.18459294587373734,
|
|
"num_tokens": 8980141.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 6.171630430221557,
|
|
"epoch": 4.562904027846843,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034682753017142387,
|
|
"loss": 5.6334,
|
|
"mean_token_accuracy": 0.18880342543125153,
|
|
"num_tokens": 8989514.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 6.179600477218628,
|
|
"epoch": 4.567876678269518,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034645732671422517,
|
|
"loss": 5.7239,
|
|
"mean_token_accuracy": 0.17676708102226257,
|
|
"num_tokens": 9000054.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 6.2347643852233885,
|
|
"epoch": 4.572849328692193,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003460869079841809,
|
|
"loss": 5.7366,
|
|
"mean_token_accuracy": 0.18273988366127014,
|
|
"num_tokens": 9010070.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 6.369583940505981,
|
|
"epoch": 4.577821979114868,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003457162750972172,
|
|
"loss": 5.707,
|
|
"mean_token_accuracy": 0.17476152777671813,
|
|
"num_tokens": 9019785.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 6.232218647003174,
|
|
"epoch": 4.582794629537544,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034534542916990534,
|
|
"loss": 5.6967,
|
|
"mean_token_accuracy": 0.18101131319999694,
|
|
"num_tokens": 9028281.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 6.105072450637818,
|
|
"epoch": 4.587767279960219,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034497437131945837,
|
|
"loss": 5.4543,
|
|
"mean_token_accuracy": 0.2034740060567856,
|
|
"num_tokens": 9037628.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 6.325877571105957,
|
|
"epoch": 4.592739930382894,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003446031026637277,
|
|
"loss": 5.71,
|
|
"mean_token_accuracy": 0.17753665894269943,
|
|
"num_tokens": 9047939.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 6.240696763992309,
|
|
"epoch": 4.597712580805569,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034423162432119995,
|
|
"loss": 5.6488,
|
|
"mean_token_accuracy": 0.1833519846200943,
|
|
"num_tokens": 9056583.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 6.184951829910278,
|
|
"epoch": 4.602685231228245,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003438599374109934,
|
|
"loss": 5.63,
|
|
"mean_token_accuracy": 0.1801805764436722,
|
|
"num_tokens": 9067610.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 6.2672646045684814,
|
|
"epoch": 4.60765788165092,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00034348804305285466,
|
|
"loss": 5.6409,
|
|
"mean_token_accuracy": 0.18079389929771422,
|
|
"num_tokens": 9077958.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 6.251768016815186,
|
|
"epoch": 4.612630532073595,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003431159423671552,
|
|
"loss": 5.734,
|
|
"mean_token_accuracy": 0.174880687892437,
|
|
"num_tokens": 9088821.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 6.203565454483032,
|
|
"epoch": 4.61760318249627,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034274363647488826,
|
|
"loss": 5.5881,
|
|
"mean_token_accuracy": 0.1848430946469307,
|
|
"num_tokens": 9098495.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 6.275817775726319,
|
|
"epoch": 4.622575832918946,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003423711264976652,
|
|
"loss": 5.6328,
|
|
"mean_token_accuracy": 0.1835610181093216,
|
|
"num_tokens": 9107279.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 6.2716064453125,
|
|
"epoch": 4.627548483341621,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034199841355771226,
|
|
"loss": 5.6641,
|
|
"mean_token_accuracy": 0.17828137278556824,
|
|
"num_tokens": 9118182.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 6.2162230014801025,
|
|
"epoch": 4.632521133764296,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034162549877786685,
|
|
"loss": 5.6278,
|
|
"mean_token_accuracy": 0.18568768799304963,
|
|
"num_tokens": 9128299.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 6.287680292129517,
|
|
"epoch": 4.6374937841869714,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003412523832815748,
|
|
"loss": 5.7521,
|
|
"mean_token_accuracy": 0.17214326858520507,
|
|
"num_tokens": 9138097.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 6.180834436416626,
|
|
"epoch": 4.642466434609647,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003408790681928865,
|
|
"loss": 5.5532,
|
|
"mean_token_accuracy": 0.18603514283895492,
|
|
"num_tokens": 9148951.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 6.237112331390381,
|
|
"epoch": 4.647439085032322,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003405055546364535,
|
|
"loss": 5.6366,
|
|
"mean_token_accuracy": 0.18117749989032744,
|
|
"num_tokens": 9158372.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 6.121814918518067,
|
|
"epoch": 4.652411735454997,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034013184373752557,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.197100493311882,
|
|
"num_tokens": 9168308.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 6.137485218048096,
|
|
"epoch": 4.6573843858776725,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003397579366219466,
|
|
"loss": 5.5446,
|
|
"mean_token_accuracy": 0.19921163618564605,
|
|
"num_tokens": 9178200.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 6.239682674407959,
|
|
"epoch": 4.662357036300348,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003393838344161519,
|
|
"loss": 5.6679,
|
|
"mean_token_accuracy": 0.1797533929347992,
|
|
"num_tokens": 9187759.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 6.073846960067749,
|
|
"epoch": 4.667329686723024,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033900953824716443,
|
|
"loss": 5.3378,
|
|
"mean_token_accuracy": 0.21129241734743118,
|
|
"num_tokens": 9198530.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 6.1716064453125,
|
|
"epoch": 4.672302337145698,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003386350492425913,
|
|
"loss": 5.6646,
|
|
"mean_token_accuracy": 0.18314312249422074,
|
|
"num_tokens": 9207668.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 6.313389825820923,
|
|
"epoch": 4.677274987568374,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033826036853062095,
|
|
"loss": 5.7829,
|
|
"mean_token_accuracy": 0.17804693728685378,
|
|
"num_tokens": 9216856.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 6.273494529724121,
|
|
"epoch": 4.68224763799105,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033788549724001897,
|
|
"loss": 5.6545,
|
|
"mean_token_accuracy": 0.18159447759389877,
|
|
"num_tokens": 9226307.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 6.1509076118469235,
|
|
"epoch": 4.687220288413725,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033751043650012534,
|
|
"loss": 5.6372,
|
|
"mean_token_accuracy": 0.18542704582214356,
|
|
"num_tokens": 9235919.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 6.094970798492431,
|
|
"epoch": 4.6921929388364,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003371351874408506,
|
|
"loss": 5.403,
|
|
"mean_token_accuracy": 0.2042030155658722,
|
|
"num_tokens": 9245542.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 6.333126497268677,
|
|
"epoch": 4.697165589259075,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003367597511926728,
|
|
"loss": 5.7288,
|
|
"mean_token_accuracy": 0.17807173877954482,
|
|
"num_tokens": 9256146.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 6.295636749267578,
|
|
"epoch": 4.702138239681751,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00033638412888663374,
|
|
"loss": 5.7856,
|
|
"mean_token_accuracy": 0.1735198363661766,
|
|
"num_tokens": 9266694.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 6.233231782913208,
|
|
"epoch": 4.707110890104426,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003360083216543358,
|
|
"loss": 5.6777,
|
|
"mean_token_accuracy": 0.1767061769962311,
|
|
"num_tokens": 9276155.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 6.2621893882751465,
|
|
"epoch": 4.712083540527101,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003356323306279387,
|
|
"loss": 5.6554,
|
|
"mean_token_accuracy": 0.1782238706946373,
|
|
"num_tokens": 9286048.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 6.350800895690918,
|
|
"epoch": 4.717056190949776,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00033525615694015545,
|
|
"loss": 5.7869,
|
|
"mean_token_accuracy": 0.16956360042095184,
|
|
"num_tokens": 9296515.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 6.109996557235718,
|
|
"epoch": 4.722028841372452,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033487980172424955,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.19532921612262727,
|
|
"num_tokens": 9306737.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 6.2578568935394285,
|
|
"epoch": 4.727001491795127,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033450326611403136,
|
|
"loss": 5.7151,
|
|
"mean_token_accuracy": 0.17832305431365966,
|
|
"num_tokens": 9316453.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 6.263491249084472,
|
|
"epoch": 4.731974142217802,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033412655124385494,
|
|
"loss": 5.65,
|
|
"mean_token_accuracy": 0.18031612932682037,
|
|
"num_tokens": 9327422.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 6.2847034454345705,
|
|
"epoch": 4.736946792640477,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003337496582486139,
|
|
"loss": 5.6247,
|
|
"mean_token_accuracy": 0.18541795313358306,
|
|
"num_tokens": 9336890.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 6.259608268737793,
|
|
"epoch": 4.741919443063153,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033337258826373896,
|
|
"loss": 5.6942,
|
|
"mean_token_accuracy": 0.18298182040452957,
|
|
"num_tokens": 9346807.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 6.252794313430786,
|
|
"epoch": 4.746892093485828,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033299534242519355,
|
|
"loss": 5.6652,
|
|
"mean_token_accuracy": 0.18041112869977952,
|
|
"num_tokens": 9356343.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 6.270813226699829,
|
|
"epoch": 4.751864743908503,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00033261792186947133,
|
|
"loss": 5.6813,
|
|
"mean_token_accuracy": 0.1750505179166794,
|
|
"num_tokens": 9366936.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 6.29380955696106,
|
|
"epoch": 4.756837394331178,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00033224032773359203,
|
|
"loss": 5.6551,
|
|
"mean_token_accuracy": 0.1862955793738365,
|
|
"num_tokens": 9377520.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 6.267157506942749,
|
|
"epoch": 4.761810044753854,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033186256115509845,
|
|
"loss": 5.7468,
|
|
"mean_token_accuracy": 0.17930571585893632,
|
|
"num_tokens": 9386498.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 6.247949934005737,
|
|
"epoch": 4.766782695176529,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003314846232720528,
|
|
"loss": 5.6887,
|
|
"mean_token_accuracy": 0.1771506205201149,
|
|
"num_tokens": 9396153.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 6.320649290084839,
|
|
"epoch": 4.771755345599204,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033110651522303346,
|
|
"loss": 5.6918,
|
|
"mean_token_accuracy": 0.18170998990535736,
|
|
"num_tokens": 9406019.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 6.228370952606201,
|
|
"epoch": 4.776727996021879,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003307282381471314,
|
|
"loss": 5.674,
|
|
"mean_token_accuracy": 0.1794967234134674,
|
|
"num_tokens": 9416019.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 6.20267882347107,
|
|
"epoch": 4.781700646444555,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033034979318394676,
|
|
"loss": 5.6457,
|
|
"mean_token_accuracy": 0.1815674424171448,
|
|
"num_tokens": 9425874.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 6.237447309494018,
|
|
"epoch": 4.78667329686723,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032997118147358554,
|
|
"loss": 5.696,
|
|
"mean_token_accuracy": 0.1811669111251831,
|
|
"num_tokens": 9435252.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 6.287495422363281,
|
|
"epoch": 4.791645947289906,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000329592404156656,
|
|
"loss": 5.642,
|
|
"mean_token_accuracy": 0.17946914583444595,
|
|
"num_tokens": 9444696.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 6.244489145278931,
|
|
"epoch": 4.79661859771258,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032921346237426536,
|
|
"loss": 5.674,
|
|
"mean_token_accuracy": 0.1821028172969818,
|
|
"num_tokens": 9454163.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 6.203081750869751,
|
|
"epoch": 4.8015912481352565,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032883435726801635,
|
|
"loss": 5.4902,
|
|
"mean_token_accuracy": 0.19838393330574036,
|
|
"num_tokens": 9463058.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 6.231932497024536,
|
|
"epoch": 4.806563898557932,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003284550899800036,
|
|
"loss": 5.632,
|
|
"mean_token_accuracy": 0.18834817707538604,
|
|
"num_tokens": 9472302.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 6.297763633728027,
|
|
"epoch": 4.811536548980607,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032807566165281055,
|
|
"loss": 5.6996,
|
|
"mean_token_accuracy": 0.1762622117996216,
|
|
"num_tokens": 9482607.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 6.230083036422729,
|
|
"epoch": 4.816509199403282,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003276960734295054,
|
|
"loss": 5.6794,
|
|
"mean_token_accuracy": 0.18452463746070863,
|
|
"num_tokens": 9492451.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 6.223843574523926,
|
|
"epoch": 4.8214818498259575,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032731632645363856,
|
|
"loss": 5.6995,
|
|
"mean_token_accuracy": 0.17677178382873535,
|
|
"num_tokens": 9502633.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 6.24560513496399,
|
|
"epoch": 4.826454500248633,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003269364218692384,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.1954878717660904,
|
|
"num_tokens": 9512887.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 6.205993175506592,
|
|
"epoch": 4.831427150671308,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003265563608208081,
|
|
"loss": 5.6093,
|
|
"mean_token_accuracy": 0.18709066212177278,
|
|
"num_tokens": 9522210.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 6.143104124069214,
|
|
"epoch": 4.836399801093983,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003261761444533221,
|
|
"loss": 5.5265,
|
|
"mean_token_accuracy": 0.19192155450582504,
|
|
"num_tokens": 9532790.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 6.244471883773803,
|
|
"epoch": 4.8413724515166585,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003257957739122231,
|
|
"loss": 5.6461,
|
|
"mean_token_accuracy": 0.18430672436952591,
|
|
"num_tokens": 9543429.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 6.298554134368897,
|
|
"epoch": 4.846345101939334,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032541525034341805,
|
|
"loss": 5.7064,
|
|
"mean_token_accuracy": 0.18030289113521575,
|
|
"num_tokens": 9552722.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 6.223086833953857,
|
|
"epoch": 4.851317752362009,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003250345748932749,
|
|
"loss": 5.6192,
|
|
"mean_token_accuracy": 0.1877907320857048,
|
|
"num_tokens": 9560880.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 6.261195516586303,
|
|
"epoch": 4.856290402784684,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032465374870861913,
|
|
"loss": 5.6994,
|
|
"mean_token_accuracy": 0.17734375298023225,
|
|
"num_tokens": 9570553.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 6.206528282165527,
|
|
"epoch": 4.8612630532073595,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00032427277293673053,
|
|
"loss": 5.7143,
|
|
"mean_token_accuracy": 0.1780391365289688,
|
|
"num_tokens": 9581313.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 6.2692169666290285,
|
|
"epoch": 4.866235703630035,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003238916487253393,
|
|
"loss": 5.6533,
|
|
"mean_token_accuracy": 0.18263526111841202,
|
|
"num_tokens": 9589545.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 6.279321718215942,
|
|
"epoch": 4.87120835405271,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003235103772226229,
|
|
"loss": 5.6962,
|
|
"mean_token_accuracy": 0.17814692705869675,
|
|
"num_tokens": 9599489.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 6.205457067489624,
|
|
"epoch": 4.876181004475385,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003231289595772026,
|
|
"loss": 5.7399,
|
|
"mean_token_accuracy": 0.17895363122224808,
|
|
"num_tokens": 9609346.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 6.046609592437744,
|
|
"epoch": 4.8811536548980605,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003227473969381398,
|
|
"loss": 5.4021,
|
|
"mean_token_accuracy": 0.20620424151420594,
|
|
"num_tokens": 9618808.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 6.291849040985108,
|
|
"epoch": 4.886126305320736,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032236569045493283,
|
|
"loss": 5.6165,
|
|
"mean_token_accuracy": 0.18290016204118728,
|
|
"num_tokens": 9629312.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 6.285317707061767,
|
|
"epoch": 4.891098955743411,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032198384127751344,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.1733856111764908,
|
|
"num_tokens": 9639439.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 6.31062068939209,
|
|
"epoch": 4.896071606166086,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000321601850556243,
|
|
"loss": 5.6265,
|
|
"mean_token_accuracy": 0.17969216406345367,
|
|
"num_tokens": 9647824.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 6.23442325592041,
|
|
"epoch": 4.9010442565887615,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003212197194419095,
|
|
"loss": 5.6479,
|
|
"mean_token_accuracy": 0.1833755314350128,
|
|
"num_tokens": 9656962.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 6.213108015060425,
|
|
"epoch": 4.906016907011437,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003208374490857238,
|
|
"loss": 5.7152,
|
|
"mean_token_accuracy": 0.18254045844078065,
|
|
"num_tokens": 9666360.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 6.243778800964355,
|
|
"epoch": 4.910989557434112,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003204550406393164,
|
|
"loss": 5.7168,
|
|
"mean_token_accuracy": 0.1756608784198761,
|
|
"num_tokens": 9676355.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 6.266103982925415,
|
|
"epoch": 4.915962207856787,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003200724952547335,
|
|
"loss": 5.7061,
|
|
"mean_token_accuracy": 0.17769200801849366,
|
|
"num_tokens": 9686361.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 6.245471525192261,
|
|
"epoch": 4.9209348582794625,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003196898140844341,
|
|
"loss": 5.6043,
|
|
"mean_token_accuracy": 0.18604745715856552,
|
|
"num_tokens": 9696571.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 6.322285079956055,
|
|
"epoch": 4.925907508702139,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031930699828128614,
|
|
"loss": 5.8113,
|
|
"mean_token_accuracy": 0.17024898529052734,
|
|
"num_tokens": 9707595.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 6.261892938613892,
|
|
"epoch": 4.930880159124813,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003189240489985633,
|
|
"loss": 5.6995,
|
|
"mean_token_accuracy": 0.1806684210896492,
|
|
"num_tokens": 9717094.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 6.204332733154297,
|
|
"epoch": 4.935852809547489,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003185409673899412,
|
|
"loss": 5.661,
|
|
"mean_token_accuracy": 0.18079878836870195,
|
|
"num_tokens": 9727329.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 6.19036111831665,
|
|
"epoch": 4.940825459970164,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00031815775460949426,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.19158987551927567,
|
|
"num_tokens": 9736189.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 6.231839275360107,
|
|
"epoch": 4.94579811039284,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003177744118116919,
|
|
"loss": 5.6193,
|
|
"mean_token_accuracy": 0.18617125749588012,
|
|
"num_tokens": 9745760.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 6.343884038925171,
|
|
"epoch": 4.950770760815515,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003173909401513954,
|
|
"loss": 5.7743,
|
|
"mean_token_accuracy": 0.16731658354401588,
|
|
"num_tokens": 9755941.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 6.227733516693116,
|
|
"epoch": 4.95574341123819,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00031700734078385416,
|
|
"loss": 5.6196,
|
|
"mean_token_accuracy": 0.1812822476029396,
|
|
"num_tokens": 9764546.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 6.25777530670166,
|
|
"epoch": 4.960716061660865,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00031662361486470243,
|
|
"loss": 5.6812,
|
|
"mean_token_accuracy": 0.17368382215499878,
|
|
"num_tokens": 9773728.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 6.183886384963989,
|
|
"epoch": 4.965688712083541,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003162397635499555,
|
|
"loss": 5.6537,
|
|
"mean_token_accuracy": 0.1806818276643753,
|
|
"num_tokens": 9783878.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 6.2479057788848875,
|
|
"epoch": 4.970661362506216,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00031585578799600653,
|
|
"loss": 5.6814,
|
|
"mean_token_accuracy": 0.17864569127559662,
|
|
"num_tokens": 9793926.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.970661362506216,
|
|
"eval_entropy": 6.126300641468593,
|
|
"eval_loss": 6.4506120681762695,
|
|
"eval_mean_token_accuracy": 0.1465223184379015,
|
|
"eval_num_tokens": 9793926.0,
|
|
"eval_runtime": 1.9396,
|
|
"eval_samples_per_second": 1786.456,
|
|
"eval_steps_per_second": 223.758,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 6.247585010528565,
|
|
"epoch": 4.975634012928891,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00031547168935962304,
|
|
"loss": 5.6066,
|
|
"mean_token_accuracy": 0.1897353932261467,
|
|
"num_tokens": 9804213.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 6.185073471069336,
|
|
"epoch": 4.980606663351566,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003150874687979433,
|
|
"loss": 5.6593,
|
|
"mean_token_accuracy": 0.1821853592991829,
|
|
"num_tokens": 9814151.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 6.200493240356446,
|
|
"epoch": 4.985579313774242,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031470312746847284,
|
|
"loss": 5.6402,
|
|
"mean_token_accuracy": 0.18201300948858262,
|
|
"num_tokens": 9822646.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 6.077136611938476,
|
|
"epoch": 4.990551964196917,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031431866652908107,
|
|
"loss": 5.4353,
|
|
"mean_token_accuracy": 0.202784463763237,
|
|
"num_tokens": 9833204.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 6.119516754150391,
|
|
"epoch": 4.995524614619592,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003139340871379977,
|
|
"loss": 5.5612,
|
|
"mean_token_accuracy": 0.2112242743372917,
|
|
"num_tokens": 9843150.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 6.228955904642741,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.0003135493904538093,
|
|
"loss": 5.6343,
|
|
"mean_token_accuracy": 0.1816597796148724,
|
|
"num_tokens": 9850520.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 6.250870561599731,
|
|
"epoch": 5.004972650422675,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003131645776354559,
|
|
"loss": 5.4068,
|
|
"mean_token_accuracy": 0.19494504779577254,
|
|
"num_tokens": 9860876.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 6.175684976577759,
|
|
"epoch": 5.0099453008453505,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003127796498422272,
|
|
"loss": 5.3078,
|
|
"mean_token_accuracy": 0.21331357955932617,
|
|
"num_tokens": 9870944.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 6.185745000839233,
|
|
"epoch": 5.014917951268026,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00031239460823375933,
|
|
"loss": 5.3713,
|
|
"mean_token_accuracy": 0.1981504514813423,
|
|
"num_tokens": 9880615.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 6.071499252319336,
|
|
"epoch": 5.019890601690701,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00031200945397003157,
|
|
"loss": 5.1953,
|
|
"mean_token_accuracy": 0.21724689602851868,
|
|
"num_tokens": 9889933.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 6.24455246925354,
|
|
"epoch": 5.024863252113376,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003116241882113623,
|
|
"loss": 5.4566,
|
|
"mean_token_accuracy": 0.1856927305459976,
|
|
"num_tokens": 9900166.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 6.120909547805786,
|
|
"epoch": 5.0298359025360515,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00031123881211840565,
|
|
"loss": 5.3612,
|
|
"mean_token_accuracy": 0.19856953918933867,
|
|
"num_tokens": 9909747.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 6.190300226211548,
|
|
"epoch": 5.034808552958727,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003108533268521485,
|
|
"loss": 5.4238,
|
|
"mean_token_accuracy": 0.19998795092105864,
|
|
"num_tokens": 9919039.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 6.248425436019898,
|
|
"epoch": 5.039781203381402,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003104677335739065,
|
|
"loss": 5.4327,
|
|
"mean_token_accuracy": 0.2001147225499153,
|
|
"num_tokens": 9927933.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 6.2189305305480955,
|
|
"epoch": 5.044753853804077,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00031008203344532067,
|
|
"loss": 5.4297,
|
|
"mean_token_accuracy": 0.19400442242622376,
|
|
"num_tokens": 9939264.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 6.185387802124024,
|
|
"epoch": 5.0497265042267525,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000309696227628354,
|
|
"loss": 5.4499,
|
|
"mean_token_accuracy": 0.19498560577630997,
|
|
"num_tokens": 9948375.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 6.23726806640625,
|
|
"epoch": 5.054699154649428,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030931031728528754,
|
|
"loss": 5.3199,
|
|
"mean_token_accuracy": 0.20206840187311173,
|
|
"num_tokens": 9957792.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 6.212782144546509,
|
|
"epoch": 5.059671805072103,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003089243035787178,
|
|
"loss": 5.4064,
|
|
"mean_token_accuracy": 0.20197986662387848,
|
|
"num_tokens": 9967213.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 6.189104747772217,
|
|
"epoch": 5.064644455494779,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00030853818767155225,
|
|
"loss": 5.4359,
|
|
"mean_token_accuracy": 0.19544170945882797,
|
|
"num_tokens": 9976886.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 6.172565889358521,
|
|
"epoch": 5.069617105917454,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003081519707270065,
|
|
"loss": 5.3645,
|
|
"mean_token_accuracy": 0.20492349117994307,
|
|
"num_tokens": 9986702.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 6.2294951438903805,
|
|
"epoch": 5.07458975634013,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00030776565390860017,
|
|
"loss": 5.3585,
|
|
"mean_token_accuracy": 0.20937828123569488,
|
|
"num_tokens": 9996597.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 6.194379663467407,
|
|
"epoch": 5.079562406762805,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003073792383801543,
|
|
"loss": 5.4283,
|
|
"mean_token_accuracy": 0.1935900017619133,
|
|
"num_tokens": 10006440.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 6.187906980514526,
|
|
"epoch": 5.08453505718548,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00030699272530578697,
|
|
"loss": 5.4409,
|
|
"mean_token_accuracy": 0.19982813894748688,
|
|
"num_tokens": 10016198.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 6.23057713508606,
|
|
"epoch": 5.089507707608155,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030660611584991013,
|
|
"loss": 5.4566,
|
|
"mean_token_accuracy": 0.19759474396705629,
|
|
"num_tokens": 10025795.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 6.1693953514099125,
|
|
"epoch": 5.094480358030831,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00030621941117722615,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.20983651876449586,
|
|
"num_tokens": 10035448.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 6.171275329589844,
|
|
"epoch": 5.099453008453506,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003058326124527243,
|
|
"loss": 5.4176,
|
|
"mean_token_accuracy": 0.20218128710985184,
|
|
"num_tokens": 10046335.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 6.261319351196289,
|
|
"epoch": 5.104425658876181,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003054457208416771,
|
|
"loss": 5.4705,
|
|
"mean_token_accuracy": 0.19648723304271698,
|
|
"num_tokens": 10056346.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 6.060270929336548,
|
|
"epoch": 5.109398309298856,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030505873750963705,
|
|
"loss": 5.267,
|
|
"mean_token_accuracy": 0.21455192863941192,
|
|
"num_tokens": 10065948.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 6.145488023757935,
|
|
"epoch": 5.114370959721532,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0003046716636224327,
|
|
"loss": 5.3551,
|
|
"mean_token_accuracy": 0.20434091091156006,
|
|
"num_tokens": 10076680.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 6.223892736434936,
|
|
"epoch": 5.119343610144207,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003042845003461657,
|
|
"loss": 5.4043,
|
|
"mean_token_accuracy": 0.19425256699323654,
|
|
"num_tokens": 10086373.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 6.176059198379517,
|
|
"epoch": 5.124316260566882,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00030389724884720664,
|
|
"loss": 5.4133,
|
|
"mean_token_accuracy": 0.1974940299987793,
|
|
"num_tokens": 10095415.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 6.035350751876831,
|
|
"epoch": 5.129288910989557,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00030350991029219236,
|
|
"loss": 5.2678,
|
|
"mean_token_accuracy": 0.2170449137687683,
|
|
"num_tokens": 10104709.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 6.17240891456604,
|
|
"epoch": 5.134261561412233,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00030312248584802166,
|
|
"loss": 5.4128,
|
|
"mean_token_accuracy": 0.20166515111923217,
|
|
"num_tokens": 10113539.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 6.111903238296509,
|
|
"epoch": 5.139234211834908,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003027349766818521,
|
|
"loss": 5.2991,
|
|
"mean_token_accuracy": 0.21330618858337402,
|
|
"num_tokens": 10123914.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 6.202079629898071,
|
|
"epoch": 5.144206862257583,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003023473839610965,
|
|
"loss": 5.4805,
|
|
"mean_token_accuracy": 0.1915132075548172,
|
|
"num_tokens": 10133572.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 6.258308219909668,
|
|
"epoch": 5.149179512680258,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003019597088534195,
|
|
"loss": 5.5111,
|
|
"mean_token_accuracy": 0.1914551794528961,
|
|
"num_tokens": 10142973.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 6.216809463500977,
|
|
"epoch": 5.154152163102934,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003015719525267338,
|
|
"loss": 5.5032,
|
|
"mean_token_accuracy": 0.19267797768115996,
|
|
"num_tokens": 10153322.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 6.157892465591431,
|
|
"epoch": 5.159124813525609,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003011841161491969,
|
|
"loss": 5.3986,
|
|
"mean_token_accuracy": 0.19993923157453536,
|
|
"num_tokens": 10163176.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 6.185671377182007,
|
|
"epoch": 5.164097463948284,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00030079620088920727,
|
|
"loss": 5.4103,
|
|
"mean_token_accuracy": 0.19549495875835418,
|
|
"num_tokens": 10172616.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 6.230748510360717,
|
|
"epoch": 5.169070114370959,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003004082079154013,
|
|
"loss": 5.4277,
|
|
"mean_token_accuracy": 0.19644053727388383,
|
|
"num_tokens": 10182243.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 6.188282537460327,
|
|
"epoch": 5.174042764793635,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00030002013839664925,
|
|
"loss": 5.4012,
|
|
"mean_token_accuracy": 0.19823480993509293,
|
|
"num_tokens": 10191394.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.99615888595581,
|
|
"epoch": 5.17901541521631,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002996319935020522,
|
|
"loss": 5.2596,
|
|
"mean_token_accuracy": 0.22089662551879882,
|
|
"num_tokens": 10202167.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 6.174089145660401,
|
|
"epoch": 5.183988065638985,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002992437744009381,
|
|
"loss": 5.3292,
|
|
"mean_token_accuracy": 0.19758762866258622,
|
|
"num_tokens": 10211250.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 6.2892594814300535,
|
|
"epoch": 5.188960716061661,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00029885548226285855,
|
|
"loss": 5.4763,
|
|
"mean_token_accuracy": 0.1918208569288254,
|
|
"num_tokens": 10220722.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 6.18586802482605,
|
|
"epoch": 5.193933366484337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002984671182575851,
|
|
"loss": 5.4764,
|
|
"mean_token_accuracy": 0.19237081259489058,
|
|
"num_tokens": 10229880.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 6.134345674514771,
|
|
"epoch": 5.198906016907012,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002980786835551061,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.19590073376893996,
|
|
"num_tokens": 10240939.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 6.238071060180664,
|
|
"epoch": 5.203878667329687,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002976901793256225,
|
|
"loss": 5.4323,
|
|
"mean_token_accuracy": 0.19686574935913087,
|
|
"num_tokens": 10249383.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 6.314925622940064,
|
|
"epoch": 5.208851317752362,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000297301606739545,
|
|
"loss": 5.4801,
|
|
"mean_token_accuracy": 0.1946067065000534,
|
|
"num_tokens": 10259166.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 6.262564945220947,
|
|
"epoch": 5.213823968175038,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00029691296696748994,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.18938307613134384,
|
|
"num_tokens": 10269053.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 6.139421606063843,
|
|
"epoch": 5.218796618597713,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00029652426118027635,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.20470174103975297,
|
|
"num_tokens": 10279148.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 6.124644231796265,
|
|
"epoch": 5.223769269020388,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00029613549054892206,
|
|
"loss": 5.3763,
|
|
"mean_token_accuracy": 0.19695284366607665,
|
|
"num_tokens": 10288719.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 6.176375198364258,
|
|
"epoch": 5.228741919443063,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002957466562446401,
|
|
"loss": 5.3989,
|
|
"mean_token_accuracy": 0.199228698015213,
|
|
"num_tokens": 10300089.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 6.217576122283935,
|
|
"epoch": 5.233714569865739,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002953577594388355,
|
|
"loss": 5.4061,
|
|
"mean_token_accuracy": 0.1960652396082878,
|
|
"num_tokens": 10309400.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 6.2070564270019535,
|
|
"epoch": 5.238687220288414,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00029496880130310145,
|
|
"loss": 5.4689,
|
|
"mean_token_accuracy": 0.19784503281116486,
|
|
"num_tokens": 10319115.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 6.191805076599121,
|
|
"epoch": 5.243659870711089,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002945797830092161,
|
|
"loss": 5.3541,
|
|
"mean_token_accuracy": 0.19632308483123778,
|
|
"num_tokens": 10327174.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 6.220529413223266,
|
|
"epoch": 5.248632521133764,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002941907057291386,
|
|
"loss": 5.4459,
|
|
"mean_token_accuracy": 0.19678962230682373,
|
|
"num_tokens": 10335608.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 6.124510097503662,
|
|
"epoch": 5.25360517155644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00029380157063500594,
|
|
"loss": 5.3316,
|
|
"mean_token_accuracy": 0.20623367875814438,
|
|
"num_tokens": 10345592.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 6.083023262023926,
|
|
"epoch": 5.258577821979115,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002934123788991291,
|
|
"loss": 5.3612,
|
|
"mean_token_accuracy": 0.20763802975416185,
|
|
"num_tokens": 10355913.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 6.163168811798096,
|
|
"epoch": 5.26355047240179,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00029302313169399,
|
|
"loss": 5.4473,
|
|
"mean_token_accuracy": 0.19477636069059373,
|
|
"num_tokens": 10365419.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 6.217544031143189,
|
|
"epoch": 5.268523122824465,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002926338301922376,
|
|
"loss": 5.4572,
|
|
"mean_token_accuracy": 0.19125280827283858,
|
|
"num_tokens": 10373939.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 6.178269290924073,
|
|
"epoch": 5.273495773247141,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00029224447556668417,
|
|
"loss": 5.4951,
|
|
"mean_token_accuracy": 0.18438511192798615,
|
|
"num_tokens": 10383932.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 6.225351238250733,
|
|
"epoch": 5.278468423669816,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002918550689903022,
|
|
"loss": 5.4228,
|
|
"mean_token_accuracy": 0.19614978581666948,
|
|
"num_tokens": 10393319.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 6.159946537017822,
|
|
"epoch": 5.283441074092491,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00029146561163622076,
|
|
"loss": 5.5427,
|
|
"mean_token_accuracy": 0.18265835493803023,
|
|
"num_tokens": 10404102.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 6.136913251876831,
|
|
"epoch": 5.288413724515166,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002910761046777218,
|
|
"loss": 5.4014,
|
|
"mean_token_accuracy": 0.19864488393068314,
|
|
"num_tokens": 10412718.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 6.240938568115235,
|
|
"epoch": 5.293386374937842,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029068654928823666,
|
|
"loss": 5.5004,
|
|
"mean_token_accuracy": 0.1902594670653343,
|
|
"num_tokens": 10422546.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 6.2006597995758055,
|
|
"epoch": 5.298359025360517,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00029029694664134284,
|
|
"loss": 5.4826,
|
|
"mean_token_accuracy": 0.1922459989786148,
|
|
"num_tokens": 10431498.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 6.166826868057251,
|
|
"epoch": 5.303331675783192,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002899072979107597,
|
|
"loss": 5.4819,
|
|
"mean_token_accuracy": 0.19203427582979202,
|
|
"num_tokens": 10441713.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 6.235426425933838,
|
|
"epoch": 5.308304326205867,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002895176042703461,
|
|
"loss": 5.4915,
|
|
"mean_token_accuracy": 0.1898449629545212,
|
|
"num_tokens": 10451570.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 6.223248529434204,
|
|
"epoch": 5.3132769766285435,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00028912786689409554,
|
|
"loss": 5.4665,
|
|
"mean_token_accuracy": 0.19739556163549424,
|
|
"num_tokens": 10461442.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 6.239466667175293,
|
|
"epoch": 5.318249627051218,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002887380869561338,
|
|
"loss": 5.5309,
|
|
"mean_token_accuracy": 0.1892786368727684,
|
|
"num_tokens": 10471223.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 6.178134202957153,
|
|
"epoch": 5.323222277473894,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002883482656307145,
|
|
"loss": 5.3854,
|
|
"mean_token_accuracy": 0.19344528913497924,
|
|
"num_tokens": 10480469.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 6.147064256668091,
|
|
"epoch": 5.328194927896569,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002879584040922162,
|
|
"loss": 5.4578,
|
|
"mean_token_accuracy": 0.19626338481903077,
|
|
"num_tokens": 10490320.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 6.154359149932861,
|
|
"epoch": 5.3331675783192445,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002875685035151386,
|
|
"loss": 5.4684,
|
|
"mean_token_accuracy": 0.19108934104442596,
|
|
"num_tokens": 10500471.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 6.258134937286377,
|
|
"epoch": 5.33814022874192,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00028717856507409886,
|
|
"loss": 5.5576,
|
|
"mean_token_accuracy": 0.1817059889435768,
|
|
"num_tokens": 10510956.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 6.257118511199951,
|
|
"epoch": 5.343112879164595,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002867885899438281,
|
|
"loss": 5.5111,
|
|
"mean_token_accuracy": 0.19526046216487886,
|
|
"num_tokens": 10520838.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 6.245999431610107,
|
|
"epoch": 5.34808552958727,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002863985792991683,
|
|
"loss": 5.4693,
|
|
"mean_token_accuracy": 0.19325918555259705,
|
|
"num_tokens": 10530879.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 6.065944719314575,
|
|
"epoch": 5.3530581800099455,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002860085343150683,
|
|
"loss": 5.2995,
|
|
"mean_token_accuracy": 0.2177024856209755,
|
|
"num_tokens": 10540161.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 6.200138902664184,
|
|
"epoch": 5.358030830432621,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028561845616658033,
|
|
"loss": 5.4815,
|
|
"mean_token_accuracy": 0.19535797387361525,
|
|
"num_tokens": 10551173.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 6.201765441894532,
|
|
"epoch": 5.363003480855296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00028522834602885656,
|
|
"loss": 5.4232,
|
|
"mean_token_accuracy": 0.1927090674638748,
|
|
"num_tokens": 10560882.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 6.238380002975464,
|
|
"epoch": 5.367976131277971,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002848382050771455,
|
|
"loss": 5.5628,
|
|
"mean_token_accuracy": 0.18690551966428756,
|
|
"num_tokens": 10570837.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 6.205521774291992,
|
|
"epoch": 5.3729487817006465,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002844480344867886,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.19858883172273636,
|
|
"num_tokens": 10580335.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 6.176025867462158,
|
|
"epoch": 5.377921432123322,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002840578354332165,
|
|
"loss": 5.4616,
|
|
"mean_token_accuracy": 0.19575107395648955,
|
|
"num_tokens": 10589030.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 6.148245000839234,
|
|
"epoch": 5.382894082545997,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002836676090919456,
|
|
"loss": 5.4473,
|
|
"mean_token_accuracy": 0.19040344506502152,
|
|
"num_tokens": 10599309.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 6.2050333499908445,
|
|
"epoch": 5.387866732968672,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002832773566385746,
|
|
"loss": 5.4102,
|
|
"mean_token_accuracy": 0.197524356842041,
|
|
"num_tokens": 10608781.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 6.177842235565185,
|
|
"epoch": 5.3928393833913475,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00028288707924878056,
|
|
"loss": 5.3594,
|
|
"mean_token_accuracy": 0.20171034783124925,
|
|
"num_tokens": 10618512.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 6.162240362167358,
|
|
"epoch": 5.397812033814023,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002824967780983162,
|
|
"loss": 5.4747,
|
|
"mean_token_accuracy": 0.19359998404979706,
|
|
"num_tokens": 10629420.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 6.208509826660157,
|
|
"epoch": 5.402784684236698,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00028210645436300524,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.19414982795715333,
|
|
"num_tokens": 10639625.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 6.169268035888672,
|
|
"epoch": 5.407757334659373,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028171610921873995,
|
|
"loss": 5.3938,
|
|
"mean_token_accuracy": 0.1962206020951271,
|
|
"num_tokens": 10648633.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 6.066241836547851,
|
|
"epoch": 5.4127299850820485,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002813257438414766,
|
|
"loss": 5.2992,
|
|
"mean_token_accuracy": 0.2078951969742775,
|
|
"num_tokens": 10657981.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 6.106432676315308,
|
|
"epoch": 5.417702635504724,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000280935359407233,
|
|
"loss": 5.3201,
|
|
"mean_token_accuracy": 0.20523831099271775,
|
|
"num_tokens": 10669734.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 6.133610486984253,
|
|
"epoch": 5.422675285927399,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002805449570920838,
|
|
"loss": 5.4212,
|
|
"mean_token_accuracy": 0.19180251508951188,
|
|
"num_tokens": 10679486.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 6.197976922988891,
|
|
"epoch": 5.427647936350074,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00028015453807215786,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.19378553032875062,
|
|
"num_tokens": 10689551.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 6.188831949234009,
|
|
"epoch": 5.4326205867727495,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00027976410352363416,
|
|
"loss": 5.4345,
|
|
"mean_token_accuracy": 0.1907230168581009,
|
|
"num_tokens": 10699173.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 6.089778757095337,
|
|
"epoch": 5.437593237195425,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027937365462273864,
|
|
"loss": 5.3931,
|
|
"mean_token_accuracy": 0.20001833289861679,
|
|
"num_tokens": 10709050.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 6.1721296310424805,
|
|
"epoch": 5.4425658876181,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00027898319254574043,
|
|
"loss": 5.4847,
|
|
"mean_token_accuracy": 0.19324643313884735,
|
|
"num_tokens": 10718397.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 6.219713163375855,
|
|
"epoch": 5.447538538040776,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002785927184689483,
|
|
"loss": 5.5045,
|
|
"mean_token_accuracy": 0.18945570290088654,
|
|
"num_tokens": 10729305.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 6.207805109024048,
|
|
"epoch": 5.452511188463451,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002782022335687073,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.19094609320163727,
|
|
"num_tokens": 10738602.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 6.146973133087158,
|
|
"epoch": 5.457483838886127,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00027781173902139485,
|
|
"loss": 5.4207,
|
|
"mean_token_accuracy": 0.20681828409433364,
|
|
"num_tokens": 10748373.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 6.200698661804199,
|
|
"epoch": 5.462456489308802,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002774212360034176,
|
|
"loss": 5.4585,
|
|
"mean_token_accuracy": 0.192893947660923,
|
|
"num_tokens": 10756879.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 6.1107090473175045,
|
|
"epoch": 5.467429139731477,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002770307256912078,
|
|
"loss": 5.3205,
|
|
"mean_token_accuracy": 0.20352033525705338,
|
|
"num_tokens": 10766969.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 5.467429139731477,
|
|
"eval_entropy": 6.020764522288801,
|
|
"eval_loss": 6.469359874725342,
|
|
"eval_mean_token_accuracy": 0.14828395893290844,
|
|
"eval_num_tokens": 10766969.0,
|
|
"eval_runtime": 1.9322,
|
|
"eval_samples_per_second": 1793.304,
|
|
"eval_steps_per_second": 224.616,
|
|
"step": 5500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10050,
|
|
"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": 2614990792089600.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|