ec2fe396ab
Model: fpadovani/tur-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.4912978617603183,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 715803608033280.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|