2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 0.055955012170215146,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1000,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 10.691505527496338,
|
||
|
|
"epoch": 0.0002797750608510757,
|
||
|
|
"grad_norm": 12.375,
|
||
|
|
"learning_rate": 2e-06,
|
||
|
|
"loss": 10.7862,
|
||
|
|
"mean_token_accuracy": 0.0,
|
||
|
|
"num_tokens": 9833.0,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.691538619995118,
|
||
|
|
"epoch": 0.0005595501217021514,
|
||
|
|
"grad_norm": 13.0,
|
||
|
|
"learning_rate": 4.5e-06,
|
||
|
|
"loss": 10.7246,
|
||
|
|
"mean_token_accuracy": 0.0,
|
||
|
|
"num_tokens": 19666.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.691145992279052,
|
||
|
|
"epoch": 0.0008393251825532272,
|
||
|
|
"grad_norm": 9.3125,
|
||
|
|
"learning_rate": 7e-06,
|
||
|
|
"loss": 10.4747,
|
||
|
|
"mean_token_accuracy": 0.019643833697773515,
|
||
|
|
"num_tokens": 28771.0,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.682477760314942,
|
||
|
|
"epoch": 0.0011191002434043028,
|
||
|
|
"grad_norm": 6.21875,
|
||
|
|
"learning_rate": 9.5e-06,
|
||
|
|
"loss": 10.1639,
|
||
|
|
"mean_token_accuracy": 0.04193656481802464,
|
||
|
|
"num_tokens": 38625.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.617809772491455,
|
||
|
|
"epoch": 0.0013988753042553786,
|
||
|
|
"grad_norm": 4.09375,
|
||
|
|
"learning_rate": 1.2e-05,
|
||
|
|
"loss": 9.8606,
|
||
|
|
"mean_token_accuracy": 0.04725950676947832,
|
||
|
|
"num_tokens": 47864.0,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.55274839401245,
|
||
|
|
"epoch": 0.0016786503651064545,
|
||
|
|
"grad_norm": 3.421875,
|
||
|
|
"learning_rate": 1.4500000000000002e-05,
|
||
|
|
"loss": 9.6963,
|
||
|
|
"mean_token_accuracy": 0.04289307370781899,
|
||
|
|
"num_tokens": 55926.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.584246730804443,
|
||
|
|
"epoch": 0.0019584254259575303,
|
||
|
|
"grad_norm": 3.078125,
|
||
|
|
"learning_rate": 1.7000000000000003e-05,
|
||
|
|
"loss": 9.6414,
|
||
|
|
"mean_token_accuracy": 0.044558577984571454,
|
||
|
|
"num_tokens": 64915.0,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.56161117553711,
|
||
|
|
"epoch": 0.0022382004868086057,
|
||
|
|
"grad_norm": 2.65625,
|
||
|
|
"learning_rate": 1.95e-05,
|
||
|
|
"loss": 9.6116,
|
||
|
|
"mean_token_accuracy": 0.041977206617593764,
|
||
|
|
"num_tokens": 73813.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.549141788482666,
|
||
|
|
"epoch": 0.0025179755476596815,
|
||
|
|
"grad_norm": 2.421875,
|
||
|
|
"learning_rate": 2.2e-05,
|
||
|
|
"loss": 9.5772,
|
||
|
|
"mean_token_accuracy": 0.04488353617489338,
|
||
|
|
"num_tokens": 83706.0,
|
||
|
|
"step": 45
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.555339050292968,
|
||
|
|
"epoch": 0.0027977506085107573,
|
||
|
|
"grad_norm": 2.53125,
|
||
|
|
"learning_rate": 2.4500000000000003e-05,
|
||
|
|
"loss": 9.5206,
|
||
|
|
"mean_token_accuracy": 0.041607250832021235,
|
||
|
|
"num_tokens": 92739.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.562696361541748,
|
||
|
|
"epoch": 0.003077525669361833,
|
||
|
|
"grad_norm": 2.65625,
|
||
|
|
"learning_rate": 2.7e-05,
|
||
|
|
"loss": 9.4731,
|
||
|
|
"mean_token_accuracy": 0.04852877669036389,
|
||
|
|
"num_tokens": 101880.0,
|
||
|
|
"step": 55
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.513782119750976,
|
||
|
|
"epoch": 0.003357300730212909,
|
||
|
|
"grad_norm": 2.59375,
|
||
|
|
"learning_rate": 2.95e-05,
|
||
|
|
"loss": 9.4135,
|
||
|
|
"mean_token_accuracy": 0.05527102164924145,
|
||
|
|
"num_tokens": 111303.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.410881614685058,
|
||
|
|
"epoch": 0.0036370757910639847,
|
||
|
|
"grad_norm": 2.578125,
|
||
|
|
"learning_rate": 3.2e-05,
|
||
|
|
"loss": 9.2608,
|
||
|
|
"mean_token_accuracy": 0.059278345108032225,
|
||
|
|
"num_tokens": 121022.0,
|
||
|
|
"step": 65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.35908842086792,
|
||
|
|
"epoch": 0.0039168508519150606,
|
||
|
|
"grad_norm": 2.3125,
|
||
|
|
"learning_rate": 3.4500000000000005e-05,
|
||
|
|
"loss": 9.1931,
|
||
|
|
"mean_token_accuracy": 0.05288665182888508,
|
||
|
|
"num_tokens": 129764.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.414530277252197,
|
||
|
|
"epoch": 0.004196625912766136,
|
||
|
|
"grad_norm": 2.671875,
|
||
|
|
"learning_rate": 3.7e-05,
|
||
|
|
"loss": 9.12,
|
||
|
|
"mean_token_accuracy": 0.05712716057896614,
|
||
|
|
"num_tokens": 139543.0,
|
||
|
|
"step": 75
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.420390224456787,
|
||
|
|
"epoch": 0.004476400973617211,
|
||
|
|
"grad_norm": 2.453125,
|
||
|
|
"learning_rate": 3.95e-05,
|
||
|
|
"loss": 9.0894,
|
||
|
|
"mean_token_accuracy": 0.055438223481178286,
|
||
|
|
"num_tokens": 148772.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.439853763580322,
|
||
|
|
"epoch": 0.004756176034468288,
|
||
|
|
"grad_norm": 2.53125,
|
||
|
|
"learning_rate": 4.2000000000000004e-05,
|
||
|
|
"loss": 8.8746,
|
||
|
|
"mean_token_accuracy": 0.06480722092092037,
|
||
|
|
"num_tokens": 158786.0,
|
||
|
|
"step": 85
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.333017063140868,
|
||
|
|
"epoch": 0.005035951095319363,
|
||
|
|
"grad_norm": 2.46875,
|
||
|
|
"learning_rate": 4.45e-05,
|
||
|
|
"loss": 8.7978,
|
||
|
|
"mean_token_accuracy": 0.06817127540707588,
|
||
|
|
"num_tokens": 166989.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.257375049591065,
|
||
|
|
"epoch": 0.005315726156170439,
|
||
|
|
"grad_norm": 2.421875,
|
||
|
|
"learning_rate": 4.7000000000000004e-05,
|
||
|
|
"loss": 8.6945,
|
||
|
|
"mean_token_accuracy": 0.06792460419237614,
|
||
|
|
"num_tokens": 175840.0,
|
||
|
|
"step": 95
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.24816074371338,
|
||
|
|
"epoch": 0.005595501217021515,
|
||
|
|
"grad_norm": 2.484375,
|
||
|
|
"learning_rate": 4.9500000000000004e-05,
|
||
|
|
"loss": 8.6247,
|
||
|
|
"mean_token_accuracy": 0.06662830822169781,
|
||
|
|
"num_tokens": 185375.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.16532497406006,
|
||
|
|
"epoch": 0.005875276277872591,
|
||
|
|
"grad_norm": 2.125,
|
||
|
|
"learning_rate": 5.2e-05,
|
||
|
|
"loss": 8.5206,
|
||
|
|
"mean_token_accuracy": 0.07044463530182839,
|
||
|
|
"num_tokens": 194647.0,
|
||
|
|
"step": 105
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.170048713684082,
|
||
|
|
"epoch": 0.006155051338723666,
|
||
|
|
"grad_norm": 2.421875,
|
||
|
|
"learning_rate": 5.45e-05,
|
||
|
|
"loss": 8.4143,
|
||
|
|
"mean_token_accuracy": 0.07247264273464679,
|
||
|
|
"num_tokens": 203882.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.117275524139405,
|
||
|
|
"epoch": 0.006434826399574742,
|
||
|
|
"grad_norm": 2.796875,
|
||
|
|
"learning_rate": 5.7e-05,
|
||
|
|
"loss": 8.2929,
|
||
|
|
"mean_token_accuracy": 0.06882111690938472,
|
||
|
|
"num_tokens": 212641.0,
|
||
|
|
"step": 115
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 10.047082424163818,
|
||
|
|
"epoch": 0.006714601460425818,
|
||
|
|
"grad_norm": 2.125,
|
||
|
|
"learning_rate": 5.9499999999999996e-05,
|
||
|
|
"loss": 8.1784,
|
||
|
|
"mean_token_accuracy": 0.07864802330732346,
|
||
|
|
"num_tokens": 221668.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.896892261505126,
|
||
|
|
"epoch": 0.006994376521276893,
|
||
|
|
"grad_norm": 2.390625,
|
||
|
|
"learning_rate": 6.2e-05,
|
||
|
|
"loss": 8.0655,
|
||
|
|
"mean_token_accuracy": 0.07759866937994957,
|
||
|
|
"num_tokens": 230277.0,
|
||
|
|
"step": 125
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.822550010681152,
|
||
|
|
"epoch": 0.0072741515821279695,
|
||
|
|
"grad_norm": 1.8125,
|
||
|
|
"learning_rate": 6.450000000000001e-05,
|
||
|
|
"loss": 7.9853,
|
||
|
|
"mean_token_accuracy": 0.07847488112747669,
|
||
|
|
"num_tokens": 240191.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.653499507904053,
|
||
|
|
"epoch": 0.007553926642979045,
|
||
|
|
"grad_norm": 2.40625,
|
||
|
|
"learning_rate": 6.7e-05,
|
||
|
|
"loss": 7.8597,
|
||
|
|
"mean_token_accuracy": 0.08022317960858345,
|
||
|
|
"num_tokens": 249600.0,
|
||
|
|
"step": 135
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.487109375,
|
||
|
|
"epoch": 0.007833701703830121,
|
||
|
|
"grad_norm": 1.96875,
|
||
|
|
"learning_rate": 6.950000000000001e-05,
|
||
|
|
"loss": 7.7858,
|
||
|
|
"mean_token_accuracy": 0.07919244170188904,
|
||
|
|
"num_tokens": 258766.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.301309394836426,
|
||
|
|
"epoch": 0.008113476764681197,
|
||
|
|
"grad_norm": 1.84375,
|
||
|
|
"learning_rate": 7.2e-05,
|
||
|
|
"loss": 7.5919,
|
||
|
|
"mean_token_accuracy": 0.08380925506353379,
|
||
|
|
"num_tokens": 268399.0,
|
||
|
|
"step": 145
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 9.010336112976074,
|
||
|
|
"epoch": 0.008393251825532272,
|
||
|
|
"grad_norm": 1.3359375,
|
||
|
|
"learning_rate": 7.45e-05,
|
||
|
|
"loss": 7.5475,
|
||
|
|
"mean_token_accuracy": 0.08445582017302514,
|
||
|
|
"num_tokens": 278348.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.899579620361328,
|
||
|
|
"epoch": 0.008673026886383347,
|
||
|
|
"grad_norm": 1.7265625,
|
||
|
|
"learning_rate": 7.7e-05,
|
||
|
|
"loss": 7.5228,
|
||
|
|
"mean_token_accuracy": 0.07803246155381202,
|
||
|
|
"num_tokens": 288429.0,
|
||
|
|
"step": 155
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.617181396484375,
|
||
|
|
"epoch": 0.008952801947234423,
|
||
|
|
"grad_norm": 1.5078125,
|
||
|
|
"learning_rate": 7.950000000000001e-05,
|
||
|
|
"loss": 7.319,
|
||
|
|
"mean_token_accuracy": 0.09034765139222145,
|
||
|
|
"num_tokens": 297936.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.410243892669678,
|
||
|
|
"epoch": 0.0092325770080855,
|
||
|
|
"grad_norm": 1.6640625,
|
||
|
|
"learning_rate": 8.2e-05,
|
||
|
|
"loss": 7.3897,
|
||
|
|
"mean_token_accuracy": 0.08498694151639938,
|
||
|
|
"num_tokens": 307390.0,
|
||
|
|
"step": 165
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.320203971862792,
|
||
|
|
"epoch": 0.009512352068936575,
|
||
|
|
"grad_norm": 1.4375,
|
||
|
|
"learning_rate": 8.450000000000001e-05,
|
||
|
|
"loss": 7.3416,
|
||
|
|
"mean_token_accuracy": 0.07727829068899154,
|
||
|
|
"num_tokens": 316381.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.236638164520263,
|
||
|
|
"epoch": 0.00979212712978765,
|
||
|
|
"grad_norm": 1.96875,
|
||
|
|
"learning_rate": 8.7e-05,
|
||
|
|
"loss": 7.3116,
|
||
|
|
"mean_token_accuracy": 0.07726738080382348,
|
||
|
|
"num_tokens": 325086.0,
|
||
|
|
"step": 175
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 8.076795291900634,
|
||
|
|
"epoch": 0.010071902190638726,
|
||
|
|
"grad_norm": 1.5546875,
|
||
|
|
"learning_rate": 8.95e-05,
|
||
|
|
"loss": 7.1784,
|
||
|
|
"mean_token_accuracy": 0.08438889384269714,
|
||
|
|
"num_tokens": 334438.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.889586877822876,
|
||
|
|
"epoch": 0.010351677251489803,
|
||
|
|
"grad_norm": 1.3515625,
|
||
|
|
"learning_rate": 9.2e-05,
|
||
|
|
"loss": 7.1976,
|
||
|
|
"mean_token_accuracy": 0.0870728187263012,
|
||
|
|
"num_tokens": 343169.0,
|
||
|
|
"step": 185
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.912118721008301,
|
||
|
|
"epoch": 0.010631452312340878,
|
||
|
|
"grad_norm": 1.734375,
|
||
|
|
"learning_rate": 9.45e-05,
|
||
|
|
"loss": 7.126,
|
||
|
|
"mean_token_accuracy": 0.08688639178872108,
|
||
|
|
"num_tokens": 352260.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.820125246047974,
|
||
|
|
"epoch": 0.010911227373191954,
|
||
|
|
"grad_norm": 1.6796875,
|
||
|
|
"learning_rate": 9.7e-05,
|
||
|
|
"loss": 7.0827,
|
||
|
|
"mean_token_accuracy": 0.08836827799677849,
|
||
|
|
"num_tokens": 361436.0,
|
||
|
|
"step": 195
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.675041007995605,
|
||
|
|
"epoch": 0.01119100243404303,
|
||
|
|
"grad_norm": 1.3515625,
|
||
|
|
"learning_rate": 9.95e-05,
|
||
|
|
"loss": 7.1613,
|
||
|
|
"mean_token_accuracy": 0.09058133289217948,
|
||
|
|
"num_tokens": 370232.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.803629922866821,
|
||
|
|
"epoch": 0.011470777494894105,
|
||
|
|
"grad_norm": 1.5234375,
|
||
|
|
"learning_rate": 0.000102,
|
||
|
|
"loss": 7.1257,
|
||
|
|
"mean_token_accuracy": 0.0905453845858574,
|
||
|
|
"num_tokens": 380211.0,
|
||
|
|
"step": 205
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.76220006942749,
|
||
|
|
"epoch": 0.011750552555745182,
|
||
|
|
"grad_norm": 1.390625,
|
||
|
|
"learning_rate": 0.00010449999999999999,
|
||
|
|
"loss": 7.1154,
|
||
|
|
"mean_token_accuracy": 0.08261686339974403,
|
||
|
|
"num_tokens": 390089.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.6790376663208,
|
||
|
|
"epoch": 0.012030327616596257,
|
||
|
|
"grad_norm": 1.734375,
|
||
|
|
"learning_rate": 0.000107,
|
||
|
|
"loss": 7.0364,
|
||
|
|
"mean_token_accuracy": 0.08655178025364876,
|
||
|
|
"num_tokens": 398513.0,
|
||
|
|
"step": 215
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.6753379821777346,
|
||
|
|
"epoch": 0.012310102677447332,
|
||
|
|
"grad_norm": 1.484375,
|
||
|
|
"learning_rate": 0.0001095,
|
||
|
|
"loss": 7.1785,
|
||
|
|
"mean_token_accuracy": 0.08271857276558876,
|
||
|
|
"num_tokens": 408114.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.658677101135254,
|
||
|
|
"epoch": 0.012589877738298408,
|
||
|
|
"grad_norm": 1.4296875,
|
||
|
|
"learning_rate": 0.000112,
|
||
|
|
"loss": 7.0324,
|
||
|
|
"mean_token_accuracy": 0.08730659186840058,
|
||
|
|
"num_tokens": 417680.0,
|
||
|
|
"step": 225
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.590711975097657,
|
||
|
|
"epoch": 0.012869652799149483,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.0001145,
|
||
|
|
"loss": 6.94,
|
||
|
|
"mean_token_accuracy": 0.0981583096086979,
|
||
|
|
"num_tokens": 426056.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.538638401031494,
|
||
|
|
"epoch": 0.01314942786000056,
|
||
|
|
"grad_norm": 1.4296875,
|
||
|
|
"learning_rate": 0.00011700000000000001,
|
||
|
|
"loss": 6.8994,
|
||
|
|
"mean_token_accuracy": 0.09497818350791931,
|
||
|
|
"num_tokens": 435465.0,
|
||
|
|
"step": 235
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.547599267959595,
|
||
|
|
"epoch": 0.013429202920851636,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.00011949999999999999,
|
||
|
|
"loss": 6.9882,
|
||
|
|
"mean_token_accuracy": 0.09514376819133759,
|
||
|
|
"num_tokens": 444295.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.532002258300781,
|
||
|
|
"epoch": 0.013708977981702711,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 0.000122,
|
||
|
|
"loss": 6.9084,
|
||
|
|
"mean_token_accuracy": 0.09089445620775223,
|
||
|
|
"num_tokens": 453111.0,
|
||
|
|
"step": 245
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.416005039215088,
|
||
|
|
"epoch": 0.013988753042553786,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 0.0001245,
|
||
|
|
"loss": 6.8671,
|
||
|
|
"mean_token_accuracy": 0.09376973509788514,
|
||
|
|
"num_tokens": 462438.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.44201397895813,
|
||
|
|
"epoch": 0.014268528103404862,
|
||
|
|
"grad_norm": 1.5234375,
|
||
|
|
"learning_rate": 0.000127,
|
||
|
|
"loss": 6.8518,
|
||
|
|
"mean_token_accuracy": 0.0919196330010891,
|
||
|
|
"num_tokens": 471610.0,
|
||
|
|
"step": 255
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.486415004730224,
|
||
|
|
"epoch": 0.014548303164255939,
|
||
|
|
"grad_norm": 1.4140625,
|
||
|
|
"learning_rate": 0.0001295,
|
||
|
|
"loss": 6.9342,
|
||
|
|
"mean_token_accuracy": 0.09321872368454934,
|
||
|
|
"num_tokens": 480849.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.40749454498291,
|
||
|
|
"epoch": 0.014828078225107014,
|
||
|
|
"grad_norm": 1.34375,
|
||
|
|
"learning_rate": 0.000132,
|
||
|
|
"loss": 6.9376,
|
||
|
|
"mean_token_accuracy": 0.08951603546738625,
|
||
|
|
"num_tokens": 489756.0,
|
||
|
|
"step": 265
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.461516523361206,
|
||
|
|
"epoch": 0.01510785328595809,
|
||
|
|
"grad_norm": 1.734375,
|
||
|
|
"learning_rate": 0.00013450000000000002,
|
||
|
|
"loss": 6.9217,
|
||
|
|
"mean_token_accuracy": 0.09437951892614364,
|
||
|
|
"num_tokens": 498818.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.285744953155517,
|
||
|
|
"epoch": 0.015387628346809165,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.00013700000000000002,
|
||
|
|
"loss": 6.8458,
|
||
|
|
"mean_token_accuracy": 0.09555465653538704,
|
||
|
|
"num_tokens": 508508.0,
|
||
|
|
"step": 275
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.444489002227783,
|
||
|
|
"epoch": 0.015667403407660242,
|
||
|
|
"grad_norm": 1.5078125,
|
||
|
|
"learning_rate": 0.0001395,
|
||
|
|
"loss": 7.0025,
|
||
|
|
"mean_token_accuracy": 0.09034469872713088,
|
||
|
|
"num_tokens": 518257.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.36638994216919,
|
||
|
|
"epoch": 0.015947178468511316,
|
||
|
|
"grad_norm": 1.328125,
|
||
|
|
"learning_rate": 0.00014199999999999998,
|
||
|
|
"loss": 6.9625,
|
||
|
|
"mean_token_accuracy": 0.08916445821523666,
|
||
|
|
"num_tokens": 527811.0,
|
||
|
|
"step": 285
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.358241605758667,
|
||
|
|
"epoch": 0.016226953529362393,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 0.0001445,
|
||
|
|
"loss": 6.8297,
|
||
|
|
"mean_token_accuracy": 0.09337828531861306,
|
||
|
|
"num_tokens": 536991.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.320535326004029,
|
||
|
|
"epoch": 0.016506728590213467,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.000147,
|
||
|
|
"loss": 6.9245,
|
||
|
|
"mean_token_accuracy": 0.0914350025355816,
|
||
|
|
"num_tokens": 546498.0,
|
||
|
|
"step": 295
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.351836109161377,
|
||
|
|
"epoch": 0.016786503651064544,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.0001495,
|
||
|
|
"loss": 6.8809,
|
||
|
|
"mean_token_accuracy": 0.09436944723129273,
|
||
|
|
"num_tokens": 556022.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.289101552963257,
|
||
|
|
"epoch": 0.01706627871191562,
|
||
|
|
"grad_norm": 1.40625,
|
||
|
|
"learning_rate": 0.000152,
|
||
|
|
"loss": 6.8212,
|
||
|
|
"mean_token_accuracy": 0.09640712589025498,
|
||
|
|
"num_tokens": 565392.0,
|
||
|
|
"step": 305
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.230247545242309,
|
||
|
|
"epoch": 0.017346053772766695,
|
||
|
|
"grad_norm": 1.46875,
|
||
|
|
"learning_rate": 0.00015450000000000001,
|
||
|
|
"loss": 6.8043,
|
||
|
|
"mean_token_accuracy": 0.09431554824113846,
|
||
|
|
"num_tokens": 574485.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.3728536605834964,
|
||
|
|
"epoch": 0.01762582883361777,
|
||
|
|
"grad_norm": 1.375,
|
||
|
|
"learning_rate": 0.000157,
|
||
|
|
"loss": 6.8343,
|
||
|
|
"mean_token_accuracy": 0.09452549517154693,
|
||
|
|
"num_tokens": 583149.0,
|
||
|
|
"step": 315
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.280808210372925,
|
||
|
|
"epoch": 0.017905603894468845,
|
||
|
|
"grad_norm": 1.5234375,
|
||
|
|
"learning_rate": 0.0001595,
|
||
|
|
"loss": 6.8855,
|
||
|
|
"mean_token_accuracy": 0.09358676373958588,
|
||
|
|
"num_tokens": 592876.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.064757013320923,
|
||
|
|
"epoch": 0.018185378955319922,
|
||
|
|
"grad_norm": 1.3203125,
|
||
|
|
"learning_rate": 0.000162,
|
||
|
|
"loss": 6.6685,
|
||
|
|
"mean_token_accuracy": 0.09988230094313622,
|
||
|
|
"num_tokens": 602001.0,
|
||
|
|
"step": 325
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.323857975006104,
|
||
|
|
"epoch": 0.018465154016171,
|
||
|
|
"grad_norm": 1.484375,
|
||
|
|
"learning_rate": 0.00016450000000000001,
|
||
|
|
"loss": 6.8667,
|
||
|
|
"mean_token_accuracy": 0.09093789532780647,
|
||
|
|
"num_tokens": 611602.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.089147615432739,
|
||
|
|
"epoch": 0.018744929077022073,
|
||
|
|
"grad_norm": 1.34375,
|
||
|
|
"learning_rate": 0.00016700000000000002,
|
||
|
|
"loss": 6.7506,
|
||
|
|
"mean_token_accuracy": 0.09642454162240029,
|
||
|
|
"num_tokens": 620905.0,
|
||
|
|
"step": 335
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.273478889465332,
|
||
|
|
"epoch": 0.01902470413787315,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.00016950000000000003,
|
||
|
|
"loss": 6.8279,
|
||
|
|
"mean_token_accuracy": 0.0952286258339882,
|
||
|
|
"num_tokens": 630151.0,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.070547962188721,
|
||
|
|
"epoch": 0.019304479198724227,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.00017199999999999998,
|
||
|
|
"loss": 6.7905,
|
||
|
|
"mean_token_accuracy": 0.09692409187555313,
|
||
|
|
"num_tokens": 639864.0,
|
||
|
|
"step": 345
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.4187744140625,
|
||
|
|
"epoch": 0.0195842542595753,
|
||
|
|
"grad_norm": 1.3515625,
|
||
|
|
"learning_rate": 0.00017449999999999999,
|
||
|
|
"loss": 6.8707,
|
||
|
|
"mean_token_accuracy": 0.09315617531538009,
|
||
|
|
"num_tokens": 648819.0,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.169836378097534,
|
||
|
|
"epoch": 0.019864029320426378,
|
||
|
|
"grad_norm": 1.4140625,
|
||
|
|
"learning_rate": 0.000177,
|
||
|
|
"loss": 6.8027,
|
||
|
|
"mean_token_accuracy": 0.09456639736890793,
|
||
|
|
"num_tokens": 658438.0,
|
||
|
|
"step": 355
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.143348264694214,
|
||
|
|
"epoch": 0.020143804381277452,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.0001795,
|
||
|
|
"loss": 6.7459,
|
||
|
|
"mean_token_accuracy": 0.09843514785170555,
|
||
|
|
"num_tokens": 668140.0,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.207337665557861,
|
||
|
|
"epoch": 0.02042357944212853,
|
||
|
|
"grad_norm": 1.3203125,
|
||
|
|
"learning_rate": 0.000182,
|
||
|
|
"loss": 6.7733,
|
||
|
|
"mean_token_accuracy": 0.09692016914486885,
|
||
|
|
"num_tokens": 677505.0,
|
||
|
|
"step": 365
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.047050952911377,
|
||
|
|
"epoch": 0.020703354502979606,
|
||
|
|
"grad_norm": 1.0703125,
|
||
|
|
"learning_rate": 0.0001845,
|
||
|
|
"loss": 6.7596,
|
||
|
|
"mean_token_accuracy": 0.09099765941500663,
|
||
|
|
"num_tokens": 687911.0,
|
||
|
|
"step": 370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.110365295410157,
|
||
|
|
"epoch": 0.02098312956383068,
|
||
|
|
"grad_norm": 1.40625,
|
||
|
|
"learning_rate": 0.000187,
|
||
|
|
"loss": 6.7002,
|
||
|
|
"mean_token_accuracy": 0.10557077825069427,
|
||
|
|
"num_tokens": 696565.0,
|
||
|
|
"step": 375
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.068933725357056,
|
||
|
|
"epoch": 0.021262904624681757,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.0001895,
|
||
|
|
"loss": 6.6321,
|
||
|
|
"mean_token_accuracy": 0.09727629125118256,
|
||
|
|
"num_tokens": 704918.0,
|
||
|
|
"step": 380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.053268480300903,
|
||
|
|
"epoch": 0.02154267968553283,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 0.000192,
|
||
|
|
"loss": 6.6612,
|
||
|
|
"mean_token_accuracy": 0.1000488631427288,
|
||
|
|
"num_tokens": 716172.0,
|
||
|
|
"step": 385
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.058245515823364,
|
||
|
|
"epoch": 0.021822454746383908,
|
||
|
|
"grad_norm": 1.46875,
|
||
|
|
"learning_rate": 0.0001945,
|
||
|
|
"loss": 6.6344,
|
||
|
|
"mean_token_accuracy": 0.10481962487101555,
|
||
|
|
"num_tokens": 725075.0,
|
||
|
|
"step": 390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.085121202468872,
|
||
|
|
"epoch": 0.022102229807234985,
|
||
|
|
"grad_norm": 1.4765625,
|
||
|
|
"learning_rate": 0.00019700000000000002,
|
||
|
|
"loss": 6.7861,
|
||
|
|
"mean_token_accuracy": 0.09666902050375939,
|
||
|
|
"num_tokens": 733314.0,
|
||
|
|
"step": 395
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.137383270263672,
|
||
|
|
"epoch": 0.02238200486808606,
|
||
|
|
"grad_norm": 1.359375,
|
||
|
|
"learning_rate": 0.00019950000000000002,
|
||
|
|
"loss": 6.6717,
|
||
|
|
"mean_token_accuracy": 0.10260491818189621,
|
||
|
|
"num_tokens": 742390.0,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.997572708129883,
|
||
|
|
"epoch": 0.022661779928937135,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.000202,
|
||
|
|
"loss": 6.7582,
|
||
|
|
"mean_token_accuracy": 0.09590908735990525,
|
||
|
|
"num_tokens": 751464.0,
|
||
|
|
"step": 405
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.058079195022583,
|
||
|
|
"epoch": 0.02294155498978821,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.00020449999999999998,
|
||
|
|
"loss": 6.6488,
|
||
|
|
"mean_token_accuracy": 0.10047566667199134,
|
||
|
|
"num_tokens": 760746.0,
|
||
|
|
"step": 410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.995281791687011,
|
||
|
|
"epoch": 0.023221330050639286,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.000207,
|
||
|
|
"loss": 6.5617,
|
||
|
|
"mean_token_accuracy": 0.10464712381362914,
|
||
|
|
"num_tokens": 769052.0,
|
||
|
|
"step": 415
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.989618110656738,
|
||
|
|
"epoch": 0.023501105111490363,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.0002095,
|
||
|
|
"loss": 6.7826,
|
||
|
|
"mean_token_accuracy": 0.09742124751210213,
|
||
|
|
"num_tokens": 778660.0,
|
||
|
|
"step": 420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.074799585342407,
|
||
|
|
"epoch": 0.023780880172341437,
|
||
|
|
"grad_norm": 1.296875,
|
||
|
|
"learning_rate": 0.000212,
|
||
|
|
"loss": 6.6877,
|
||
|
|
"mean_token_accuracy": 0.10435819402337074,
|
||
|
|
"num_tokens": 787841.0,
|
||
|
|
"step": 425
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.878971099853516,
|
||
|
|
"epoch": 0.024060655233192514,
|
||
|
|
"grad_norm": 1.328125,
|
||
|
|
"learning_rate": 0.0002145,
|
||
|
|
"loss": 6.537,
|
||
|
|
"mean_token_accuracy": 0.10397473350167274,
|
||
|
|
"num_tokens": 796175.0,
|
||
|
|
"step": 430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.980287361145019,
|
||
|
|
"epoch": 0.024340430294043588,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.00021700000000000002,
|
||
|
|
"loss": 6.5959,
|
||
|
|
"mean_token_accuracy": 0.10773405581712722,
|
||
|
|
"num_tokens": 805941.0,
|
||
|
|
"step": 435
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.9006028175354,
|
||
|
|
"epoch": 0.024620205354894665,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.0002195,
|
||
|
|
"loss": 6.6389,
|
||
|
|
"mean_token_accuracy": 0.09920870438218117,
|
||
|
|
"num_tokens": 815058.0,
|
||
|
|
"step": 440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.100294494628907,
|
||
|
|
"epoch": 0.024899980415745742,
|
||
|
|
"grad_norm": 1.2578125,
|
||
|
|
"learning_rate": 0.000222,
|
||
|
|
"loss": 6.6593,
|
||
|
|
"mean_token_accuracy": 0.10117991864681244,
|
||
|
|
"num_tokens": 824459.0,
|
||
|
|
"step": 445
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.872482252120972,
|
||
|
|
"epoch": 0.025179755476596816,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0002245,
|
||
|
|
"loss": 6.656,
|
||
|
|
"mean_token_accuracy": 0.10260392725467682,
|
||
|
|
"num_tokens": 835250.0,
|
||
|
|
"step": 450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.986909580230713,
|
||
|
|
"epoch": 0.025459530537447893,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 0.00022700000000000002,
|
||
|
|
"loss": 6.6617,
|
||
|
|
"mean_token_accuracy": 0.10458940342068672,
|
||
|
|
"num_tokens": 844643.0,
|
||
|
|
"step": 455
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.96947169303894,
|
||
|
|
"epoch": 0.025739305598298966,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.00022950000000000002,
|
||
|
|
"loss": 6.5977,
|
||
|
|
"mean_token_accuracy": 0.10226123109459877,
|
||
|
|
"num_tokens": 853375.0,
|
||
|
|
"step": 460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.875414848327637,
|
||
|
|
"epoch": 0.026019080659150044,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.00023200000000000003,
|
||
|
|
"loss": 6.5253,
|
||
|
|
"mean_token_accuracy": 0.11177821755409241,
|
||
|
|
"num_tokens": 862244.0,
|
||
|
|
"step": 465
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.945923662185669,
|
||
|
|
"epoch": 0.02629885572000112,
|
||
|
|
"grad_norm": 1.34375,
|
||
|
|
"learning_rate": 0.00023449999999999998,
|
||
|
|
"loss": 6.6286,
|
||
|
|
"mean_token_accuracy": 0.09996586814522743,
|
||
|
|
"num_tokens": 871828.0,
|
||
|
|
"step": 470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.8525604724884035,
|
||
|
|
"epoch": 0.026578630780852194,
|
||
|
|
"grad_norm": 1.3125,
|
||
|
|
"learning_rate": 0.000237,
|
||
|
|
"loss": 6.6076,
|
||
|
|
"mean_token_accuracy": 0.10374173521995544,
|
||
|
|
"num_tokens": 881087.0,
|
||
|
|
"step": 475
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.889187860488891,
|
||
|
|
"epoch": 0.02685840584170327,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.0002395,
|
||
|
|
"loss": 6.6779,
|
||
|
|
"mean_token_accuracy": 0.10589562878012657,
|
||
|
|
"num_tokens": 890061.0,
|
||
|
|
"step": 480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 7.009466075897217,
|
||
|
|
"epoch": 0.027138180902554345,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.000242,
|
||
|
|
"loss": 6.6192,
|
||
|
|
"mean_token_accuracy": 0.10268636345863343,
|
||
|
|
"num_tokens": 899388.0,
|
||
|
|
"step": 485
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.763098287582397,
|
||
|
|
"epoch": 0.027417955963405422,
|
||
|
|
"grad_norm": 1.3203125,
|
||
|
|
"learning_rate": 0.0002445,
|
||
|
|
"loss": 6.5255,
|
||
|
|
"mean_token_accuracy": 0.10589060559868813,
|
||
|
|
"num_tokens": 908295.0,
|
||
|
|
"step": 490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7688929557800295,
|
||
|
|
"epoch": 0.0276977310242565,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.000247,
|
||
|
|
"loss": 6.4924,
|
||
|
|
"mean_token_accuracy": 0.10882084742188454,
|
||
|
|
"num_tokens": 917514.0,
|
||
|
|
"step": 495
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.904528284072876,
|
||
|
|
"epoch": 0.027977506085107573,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0002495,
|
||
|
|
"loss": 6.5978,
|
||
|
|
"mean_token_accuracy": 0.10768755748867989,
|
||
|
|
"num_tokens": 927146.0,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.758494710922241,
|
||
|
|
"epoch": 0.02825728114595865,
|
||
|
|
"grad_norm": 1.3828125,
|
||
|
|
"learning_rate": 0.000252,
|
||
|
|
"loss": 6.4884,
|
||
|
|
"mean_token_accuracy": 0.10985259413719177,
|
||
|
|
"num_tokens": 935278.0,
|
||
|
|
"step": 505
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.831978225708008,
|
||
|
|
"epoch": 0.028537056206809724,
|
||
|
|
"grad_norm": 1.3125,
|
||
|
|
"learning_rate": 0.0002545,
|
||
|
|
"loss": 6.5894,
|
||
|
|
"mean_token_accuracy": 0.10266794636845589,
|
||
|
|
"num_tokens": 945056.0,
|
||
|
|
"step": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.827414417266846,
|
||
|
|
"epoch": 0.0288168312676608,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.000257,
|
||
|
|
"loss": 6.5527,
|
||
|
|
"mean_token_accuracy": 0.10177932679653168,
|
||
|
|
"num_tokens": 954187.0,
|
||
|
|
"step": 515
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.905342674255371,
|
||
|
|
"epoch": 0.029096606328511878,
|
||
|
|
"grad_norm": 1.140625,
|
||
|
|
"learning_rate": 0.0002595,
|
||
|
|
"loss": 6.5899,
|
||
|
|
"mean_token_accuracy": 0.10657662600278854,
|
||
|
|
"num_tokens": 963075.0,
|
||
|
|
"step": 520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.886737442016601,
|
||
|
|
"epoch": 0.02937638138936295,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000262,
|
||
|
|
"loss": 6.5683,
|
||
|
|
"mean_token_accuracy": 0.10781930312514305,
|
||
|
|
"num_tokens": 972110.0,
|
||
|
|
"step": 525
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.684601259231568,
|
||
|
|
"epoch": 0.02965615645021403,
|
||
|
|
"grad_norm": 1.3984375,
|
||
|
|
"learning_rate": 0.00026450000000000003,
|
||
|
|
"loss": 6.4582,
|
||
|
|
"mean_token_accuracy": 0.11291620060801506,
|
||
|
|
"num_tokens": 980231.0,
|
||
|
|
"step": 530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7892358779907225,
|
||
|
|
"epoch": 0.029935931511065102,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.00026700000000000004,
|
||
|
|
"loss": 6.6105,
|
||
|
|
"mean_token_accuracy": 0.10551066771149635,
|
||
|
|
"num_tokens": 990018.0,
|
||
|
|
"step": 535
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.703322696685791,
|
||
|
|
"epoch": 0.03021570657191618,
|
||
|
|
"grad_norm": 1.3203125,
|
||
|
|
"learning_rate": 0.00026950000000000005,
|
||
|
|
"loss": 6.4987,
|
||
|
|
"mean_token_accuracy": 0.10690677091479302,
|
||
|
|
"num_tokens": 999243.0,
|
||
|
|
"step": 540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.800885343551636,
|
||
|
|
"epoch": 0.030495481632767257,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.00027200000000000005,
|
||
|
|
"loss": 6.4947,
|
||
|
|
"mean_token_accuracy": 0.1084700345993042,
|
||
|
|
"num_tokens": 1008800.0,
|
||
|
|
"step": 545
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.75251727104187,
|
||
|
|
"epoch": 0.03077525669361833,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.0002745,
|
||
|
|
"loss": 6.5354,
|
||
|
|
"mean_token_accuracy": 0.10884697884321212,
|
||
|
|
"num_tokens": 1018880.0,
|
||
|
|
"step": 550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.796888923645019,
|
||
|
|
"epoch": 0.031055031754469407,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.000277,
|
||
|
|
"loss": 6.4822,
|
||
|
|
"mean_token_accuracy": 0.10811220407485962,
|
||
|
|
"num_tokens": 1028508.0,
|
||
|
|
"step": 555
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.65325436592102,
|
||
|
|
"epoch": 0.031334806815320485,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 0.0002795,
|
||
|
|
"loss": 6.4371,
|
||
|
|
"mean_token_accuracy": 0.11171221733093262,
|
||
|
|
"num_tokens": 1037626.0,
|
||
|
|
"step": 560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.6531189441680905,
|
||
|
|
"epoch": 0.031614581876171555,
|
||
|
|
"grad_norm": 1.2734375,
|
||
|
|
"learning_rate": 0.00028199999999999997,
|
||
|
|
"loss": 6.4237,
|
||
|
|
"mean_token_accuracy": 0.1133709043264389,
|
||
|
|
"num_tokens": 1046814.0,
|
||
|
|
"step": 565
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.597415113449097,
|
||
|
|
"epoch": 0.03189435693702263,
|
||
|
|
"grad_norm": 1.09375,
|
||
|
|
"learning_rate": 0.0002845,
|
||
|
|
"loss": 6.4282,
|
||
|
|
"mean_token_accuracy": 0.10929519906640053,
|
||
|
|
"num_tokens": 1056620.0,
|
||
|
|
"step": 570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.794770956039429,
|
||
|
|
"epoch": 0.03217413199787371,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.000287,
|
||
|
|
"loss": 6.4834,
|
||
|
|
"mean_token_accuracy": 0.11290100738406181,
|
||
|
|
"num_tokens": 1064488.0,
|
||
|
|
"step": 575
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.786476469039917,
|
||
|
|
"epoch": 0.032453907058724786,
|
||
|
|
"grad_norm": 1.4375,
|
||
|
|
"learning_rate": 0.0002895,
|
||
|
|
"loss": 6.5179,
|
||
|
|
"mean_token_accuracy": 0.11080160588026047,
|
||
|
|
"num_tokens": 1074153.0,
|
||
|
|
"step": 580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.673076820373535,
|
||
|
|
"epoch": 0.03273368211957586,
|
||
|
|
"grad_norm": 1.3125,
|
||
|
|
"learning_rate": 0.000292,
|
||
|
|
"loss": 6.452,
|
||
|
|
"mean_token_accuracy": 0.10839907601475715,
|
||
|
|
"num_tokens": 1083027.0,
|
||
|
|
"step": 585
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.628738451004028,
|
||
|
|
"epoch": 0.03301345718042693,
|
||
|
|
"grad_norm": 1.28125,
|
||
|
|
"learning_rate": 0.0002945,
|
||
|
|
"loss": 6.4465,
|
||
|
|
"mean_token_accuracy": 0.10820735022425651,
|
||
|
|
"num_tokens": 1092010.0,
|
||
|
|
"step": 590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.731857395172119,
|
||
|
|
"epoch": 0.03329323224127801,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.000297,
|
||
|
|
"loss": 6.4946,
|
||
|
|
"mean_token_accuracy": 0.10928055047988891,
|
||
|
|
"num_tokens": 1101776.0,
|
||
|
|
"step": 595
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.651297473907471,
|
||
|
|
"epoch": 0.03357300730212909,
|
||
|
|
"grad_norm": 1.3828125,
|
||
|
|
"learning_rate": 0.0002995,
|
||
|
|
"loss": 6.4487,
|
||
|
|
"mean_token_accuracy": 0.10809680670499802,
|
||
|
|
"num_tokens": 1110894.0,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.813410139083862,
|
||
|
|
"epoch": 0.033852782362980165,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.000302,
|
||
|
|
"loss": 6.4318,
|
||
|
|
"mean_token_accuracy": 0.10976689979434014,
|
||
|
|
"num_tokens": 1119512.0,
|
||
|
|
"step": 605
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.574000358581543,
|
||
|
|
"epoch": 0.03413255742383124,
|
||
|
|
"grad_norm": 1.265625,
|
||
|
|
"learning_rate": 0.0003045,
|
||
|
|
"loss": 6.4217,
|
||
|
|
"mean_token_accuracy": 0.10791075751185417,
|
||
|
|
"num_tokens": 1129469.0,
|
||
|
|
"step": 610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.738797283172607,
|
||
|
|
"epoch": 0.03441233248468231,
|
||
|
|
"grad_norm": 1.2578125,
|
||
|
|
"learning_rate": 0.000307,
|
||
|
|
"loss": 6.4586,
|
||
|
|
"mean_token_accuracy": 0.10540183559060097,
|
||
|
|
"num_tokens": 1137850.0,
|
||
|
|
"step": 615
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.7057750701904295,
|
||
|
|
"epoch": 0.03469210754553339,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0003095,
|
||
|
|
"loss": 6.5737,
|
||
|
|
"mean_token_accuracy": 0.11240374073386192,
|
||
|
|
"num_tokens": 1147405.0,
|
||
|
|
"step": 620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.679307174682617,
|
||
|
|
"epoch": 0.034971882606384466,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.000312,
|
||
|
|
"loss": 6.4446,
|
||
|
|
"mean_token_accuracy": 0.11059640496969222,
|
||
|
|
"num_tokens": 1156457.0,
|
||
|
|
"step": 625
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.691693830490112,
|
||
|
|
"epoch": 0.03525165766723554,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.0003145,
|
||
|
|
"loss": 6.4141,
|
||
|
|
"mean_token_accuracy": 0.11546503752470016,
|
||
|
|
"num_tokens": 1165943.0,
|
||
|
|
"step": 630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4711555480957035,
|
||
|
|
"epoch": 0.03553143272808662,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.000317,
|
||
|
|
"loss": 6.3521,
|
||
|
|
"mean_token_accuracy": 0.11288735195994377,
|
||
|
|
"num_tokens": 1175030.0,
|
||
|
|
"step": 635
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.5566987037658695,
|
||
|
|
"epoch": 0.03581120778893769,
|
||
|
|
"grad_norm": 1.2421875,
|
||
|
|
"learning_rate": 0.0003195,
|
||
|
|
"loss": 6.4719,
|
||
|
|
"mean_token_accuracy": 0.11424087956547738,
|
||
|
|
"num_tokens": 1184988.0,
|
||
|
|
"step": 640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.78665771484375,
|
||
|
|
"epoch": 0.03609098284978877,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.000322,
|
||
|
|
"loss": 6.4706,
|
||
|
|
"mean_token_accuracy": 0.11168640255928039,
|
||
|
|
"num_tokens": 1194352.0,
|
||
|
|
"step": 645
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.566987657546997,
|
||
|
|
"epoch": 0.036370757910639845,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.00032450000000000003,
|
||
|
|
"loss": 6.3646,
|
||
|
|
"mean_token_accuracy": 0.1151100791990757,
|
||
|
|
"num_tokens": 1203744.0,
|
||
|
|
"step": 650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.640985155105591,
|
||
|
|
"epoch": 0.03665053297149092,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.00032700000000000003,
|
||
|
|
"loss": 6.4265,
|
||
|
|
"mean_token_accuracy": 0.11166704073548317,
|
||
|
|
"num_tokens": 1212917.0,
|
||
|
|
"step": 655
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.689370393753052,
|
||
|
|
"epoch": 0.036930308032342,
|
||
|
|
"grad_norm": 1.328125,
|
||
|
|
"learning_rate": 0.00032950000000000004,
|
||
|
|
"loss": 6.4567,
|
||
|
|
"mean_token_accuracy": 0.11746964305639267,
|
||
|
|
"num_tokens": 1221669.0,
|
||
|
|
"step": 660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.537262868881226,
|
||
|
|
"epoch": 0.037210083093193076,
|
||
|
|
"grad_norm": 1.328125,
|
||
|
|
"learning_rate": 0.00033200000000000005,
|
||
|
|
"loss": 6.3787,
|
||
|
|
"mean_token_accuracy": 0.11580658480525016,
|
||
|
|
"num_tokens": 1230925.0,
|
||
|
|
"step": 665
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.473914861679077,
|
||
|
|
"epoch": 0.037489858154044146,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.00033450000000000005,
|
||
|
|
"loss": 6.3872,
|
||
|
|
"mean_token_accuracy": 0.12107581570744515,
|
||
|
|
"num_tokens": 1240317.0,
|
||
|
|
"step": 670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.532442855834961,
|
||
|
|
"epoch": 0.037769633214895224,
|
||
|
|
"grad_norm": 1.265625,
|
||
|
|
"learning_rate": 0.000337,
|
||
|
|
"loss": 6.3565,
|
||
|
|
"mean_token_accuracy": 0.11721108332276345,
|
||
|
|
"num_tokens": 1250285.0,
|
||
|
|
"step": 675
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.558922338485718,
|
||
|
|
"epoch": 0.0380494082757463,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.0003395,
|
||
|
|
"loss": 6.3691,
|
||
|
|
"mean_token_accuracy": 0.11794314831495285,
|
||
|
|
"num_tokens": 1259272.0,
|
||
|
|
"step": 680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.556687736511231,
|
||
|
|
"epoch": 0.03832918333659738,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.000342,
|
||
|
|
"loss": 6.3467,
|
||
|
|
"mean_token_accuracy": 0.11026200056076049,
|
||
|
|
"num_tokens": 1269069.0,
|
||
|
|
"step": 685
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.544914388656617,
|
||
|
|
"epoch": 0.038608958397448455,
|
||
|
|
"grad_norm": 1.3515625,
|
||
|
|
"learning_rate": 0.00034449999999999997,
|
||
|
|
"loss": 6.3472,
|
||
|
|
"mean_token_accuracy": 0.1141720823943615,
|
||
|
|
"num_tokens": 1277942.0,
|
||
|
|
"step": 690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.516038227081299,
|
||
|
|
"epoch": 0.038888733458299525,
|
||
|
|
"grad_norm": 1.421875,
|
||
|
|
"learning_rate": 0.000347,
|
||
|
|
"loss": 6.4464,
|
||
|
|
"mean_token_accuracy": 0.11829971224069595,
|
||
|
|
"num_tokens": 1286731.0,
|
||
|
|
"step": 695
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.491542816162109,
|
||
|
|
"epoch": 0.0391685085191506,
|
||
|
|
"grad_norm": 1.265625,
|
||
|
|
"learning_rate": 0.0003495,
|
||
|
|
"loss": 6.306,
|
||
|
|
"mean_token_accuracy": 0.11953174397349357,
|
||
|
|
"num_tokens": 1295685.0,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.41384859085083,
|
||
|
|
"epoch": 0.03944828358000168,
|
||
|
|
"grad_norm": 1.3046875,
|
||
|
|
"learning_rate": 0.000352,
|
||
|
|
"loss": 6.2558,
|
||
|
|
"mean_token_accuracy": 0.1239193856716156,
|
||
|
|
"num_tokens": 1304375.0,
|
||
|
|
"step": 705
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.570674753189087,
|
||
|
|
"epoch": 0.039728058640852756,
|
||
|
|
"grad_norm": 1.1640625,
|
||
|
|
"learning_rate": 0.0003545,
|
||
|
|
"loss": 6.4673,
|
||
|
|
"mean_token_accuracy": 0.11799883842468262,
|
||
|
|
"num_tokens": 1313282.0,
|
||
|
|
"step": 710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.577246189117432,
|
||
|
|
"epoch": 0.040007833701703834,
|
||
|
|
"grad_norm": 1.46875,
|
||
|
|
"learning_rate": 0.000357,
|
||
|
|
"loss": 6.3111,
|
||
|
|
"mean_token_accuracy": 0.11679205074906349,
|
||
|
|
"num_tokens": 1322035.0,
|
||
|
|
"step": 715
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.528969764709473,
|
||
|
|
"epoch": 0.040287608762554904,
|
||
|
|
"grad_norm": 1.359375,
|
||
|
|
"learning_rate": 0.0003595,
|
||
|
|
"loss": 6.4596,
|
||
|
|
"mean_token_accuracy": 0.11984212100505828,
|
||
|
|
"num_tokens": 1331553.0,
|
||
|
|
"step": 720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.455485153198242,
|
||
|
|
"epoch": 0.04056738382340598,
|
||
|
|
"grad_norm": 1.3828125,
|
||
|
|
"learning_rate": 0.000362,
|
||
|
|
"loss": 6.3126,
|
||
|
|
"mean_token_accuracy": 0.11634061336517335,
|
||
|
|
"num_tokens": 1340577.0,
|
||
|
|
"step": 725
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.606705093383789,
|
||
|
|
"epoch": 0.04084715888425706,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.0003645,
|
||
|
|
"loss": 6.3608,
|
||
|
|
"mean_token_accuracy": 0.12028084397315979,
|
||
|
|
"num_tokens": 1349635.0,
|
||
|
|
"step": 730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.485494136810303,
|
||
|
|
"epoch": 0.041126933945108135,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000367,
|
||
|
|
"loss": 6.2679,
|
||
|
|
"mean_token_accuracy": 0.11617309153079987,
|
||
|
|
"num_tokens": 1358628.0,
|
||
|
|
"step": 735
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.461924934387207,
|
||
|
|
"epoch": 0.04140670900595921,
|
||
|
|
"grad_norm": 1.3671875,
|
||
|
|
"learning_rate": 0.0003695,
|
||
|
|
"loss": 6.4619,
|
||
|
|
"mean_token_accuracy": 0.1150215283036232,
|
||
|
|
"num_tokens": 1368023.0,
|
||
|
|
"step": 740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4644389152526855,
|
||
|
|
"epoch": 0.04168648406681028,
|
||
|
|
"grad_norm": 1.2578125,
|
||
|
|
"learning_rate": 0.000372,
|
||
|
|
"loss": 6.338,
|
||
|
|
"mean_token_accuracy": 0.12105848267674446,
|
||
|
|
"num_tokens": 1377354.0,
|
||
|
|
"step": 745
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.552575492858887,
|
||
|
|
"epoch": 0.04196625912766136,
|
||
|
|
"grad_norm": 1.3046875,
|
||
|
|
"learning_rate": 0.0003745,
|
||
|
|
"loss": 6.4018,
|
||
|
|
"mean_token_accuracy": 0.1141360804438591,
|
||
|
|
"num_tokens": 1386428.0,
|
||
|
|
"step": 750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4100508213043215,
|
||
|
|
"epoch": 0.04224603418851244,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.000377,
|
||
|
|
"loss": 6.2951,
|
||
|
|
"mean_token_accuracy": 0.11664086356759071,
|
||
|
|
"num_tokens": 1395809.0,
|
||
|
|
"step": 755
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.542036771774292,
|
||
|
|
"epoch": 0.042525809249363514,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.0003795,
|
||
|
|
"loss": 6.322,
|
||
|
|
"mean_token_accuracy": 0.12149200588464737,
|
||
|
|
"num_tokens": 1405440.0,
|
||
|
|
"step": 760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.500925827026367,
|
||
|
|
"epoch": 0.04280558431021459,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.000382,
|
||
|
|
"loss": 6.4325,
|
||
|
|
"mean_token_accuracy": 0.11671576723456382,
|
||
|
|
"num_tokens": 1414018.0,
|
||
|
|
"step": 765
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.469825267791748,
|
||
|
|
"epoch": 0.04308535937106566,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.0003845,
|
||
|
|
"loss": 6.3286,
|
||
|
|
"mean_token_accuracy": 0.1199316494166851,
|
||
|
|
"num_tokens": 1422853.0,
|
||
|
|
"step": 770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4752647399902346,
|
||
|
|
"epoch": 0.04336513443191674,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.00038700000000000003,
|
||
|
|
"loss": 6.3173,
|
||
|
|
"mean_token_accuracy": 0.12002637311816215,
|
||
|
|
"num_tokens": 1431647.0,
|
||
|
|
"step": 775
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4410097122192385,
|
||
|
|
"epoch": 0.043644909492767815,
|
||
|
|
"grad_norm": 1.4140625,
|
||
|
|
"learning_rate": 0.00038950000000000003,
|
||
|
|
"loss": 6.3052,
|
||
|
|
"mean_token_accuracy": 0.12325269281864167,
|
||
|
|
"num_tokens": 1440547.0,
|
||
|
|
"step": 780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.460622024536133,
|
||
|
|
"epoch": 0.04392468455361889,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 0.00039200000000000004,
|
||
|
|
"loss": 6.2662,
|
||
|
|
"mean_token_accuracy": 0.12282432466745377,
|
||
|
|
"num_tokens": 1450310.0,
|
||
|
|
"step": 785
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.4368048191070555,
|
||
|
|
"epoch": 0.04420445961446997,
|
||
|
|
"grad_norm": 1.3125,
|
||
|
|
"learning_rate": 0.00039450000000000005,
|
||
|
|
"loss": 6.3488,
|
||
|
|
"mean_token_accuracy": 0.11794603615999222,
|
||
|
|
"num_tokens": 1458544.0,
|
||
|
|
"step": 790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.315302181243896,
|
||
|
|
"epoch": 0.04448423467532104,
|
||
|
|
"grad_norm": 1.25,
|
||
|
|
"learning_rate": 0.00039700000000000005,
|
||
|
|
"loss": 6.3146,
|
||
|
|
"mean_token_accuracy": 0.12033002376556397,
|
||
|
|
"num_tokens": 1467699.0,
|
||
|
|
"step": 795
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.488300848007202,
|
||
|
|
"epoch": 0.04476400973617212,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 0.0003995,
|
||
|
|
"loss": 6.2519,
|
||
|
|
"mean_token_accuracy": 0.1236931897699833,
|
||
|
|
"num_tokens": 1477386.0,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.3928173065185545,
|
||
|
|
"epoch": 0.045043784797023194,
|
||
|
|
"grad_norm": 1.4140625,
|
||
|
|
"learning_rate": 0.000402,
|
||
|
|
"loss": 6.2968,
|
||
|
|
"mean_token_accuracy": 0.11787274852395058,
|
||
|
|
"num_tokens": 1486660.0,
|
||
|
|
"step": 805
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.316296100616455,
|
||
|
|
"epoch": 0.04532355985787427,
|
||
|
|
"grad_norm": 1.2734375,
|
||
|
|
"learning_rate": 0.0004045,
|
||
|
|
"loss": 6.2154,
|
||
|
|
"mean_token_accuracy": 0.12369973957538605,
|
||
|
|
"num_tokens": 1495408.0,
|
||
|
|
"step": 810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.420361280441284,
|
||
|
|
"epoch": 0.04560333491872535,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.00040699999999999997,
|
||
|
|
"loss": 6.2966,
|
||
|
|
"mean_token_accuracy": 0.12109342366456985,
|
||
|
|
"num_tokens": 1503822.0,
|
||
|
|
"step": 815
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.339131259918213,
|
||
|
|
"epoch": 0.04588310997957642,
|
||
|
|
"grad_norm": 1.3203125,
|
||
|
|
"learning_rate": 0.0004095,
|
||
|
|
"loss": 6.2543,
|
||
|
|
"mean_token_accuracy": 0.11657608151435853,
|
||
|
|
"num_tokens": 1512822.0,
|
||
|
|
"step": 820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.436035680770874,
|
||
|
|
"epoch": 0.046162885040427495,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.000412,
|
||
|
|
"loss": 6.3429,
|
||
|
|
"mean_token_accuracy": 0.11799641102552413,
|
||
|
|
"num_tokens": 1522321.0,
|
||
|
|
"step": 825
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.388043451309204,
|
||
|
|
"epoch": 0.04644266010127857,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.0004145,
|
||
|
|
"loss": 6.3332,
|
||
|
|
"mean_token_accuracy": 0.11690129637718201,
|
||
|
|
"num_tokens": 1532071.0,
|
||
|
|
"step": 830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.446229124069214,
|
||
|
|
"epoch": 0.04672243516212965,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.000417,
|
||
|
|
"loss": 6.2658,
|
||
|
|
"mean_token_accuracy": 0.1248463362455368,
|
||
|
|
"num_tokens": 1541204.0,
|
||
|
|
"step": 835
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.321492671966553,
|
||
|
|
"epoch": 0.04700221022298073,
|
||
|
|
"grad_norm": 1.2265625,
|
||
|
|
"learning_rate": 0.0004195,
|
||
|
|
"loss": 6.301,
|
||
|
|
"mean_token_accuracy": 0.11873220205307007,
|
||
|
|
"num_tokens": 1548994.0,
|
||
|
|
"step": 840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.387647390365601,
|
||
|
|
"epoch": 0.0472819852838318,
|
||
|
|
"grad_norm": 1.390625,
|
||
|
|
"learning_rate": 0.000422,
|
||
|
|
"loss": 6.3317,
|
||
|
|
"mean_token_accuracy": 0.11347933784127236,
|
||
|
|
"num_tokens": 1558012.0,
|
||
|
|
"step": 845
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.33697543144226,
|
||
|
|
"epoch": 0.047561760344682874,
|
||
|
|
"grad_norm": 1.265625,
|
||
|
|
"learning_rate": 0.0004245,
|
||
|
|
"loss": 6.344,
|
||
|
|
"mean_token_accuracy": 0.12062845751643181,
|
||
|
|
"num_tokens": 1567183.0,
|
||
|
|
"step": 850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.517240285873413,
|
||
|
|
"epoch": 0.04784153540553395,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.000427,
|
||
|
|
"loss": 6.3582,
|
||
|
|
"mean_token_accuracy": 0.12024350762367249,
|
||
|
|
"num_tokens": 1577213.0,
|
||
|
|
"step": 855
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.327879953384399,
|
||
|
|
"epoch": 0.04812131046638503,
|
||
|
|
"grad_norm": 1.125,
|
||
|
|
"learning_rate": 0.0004295,
|
||
|
|
"loss": 6.2433,
|
||
|
|
"mean_token_accuracy": 0.12363175675272942,
|
||
|
|
"num_tokens": 1586814.0,
|
||
|
|
"step": 860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.303459978103637,
|
||
|
|
"epoch": 0.048401085527236105,
|
||
|
|
"grad_norm": 1.1796875,
|
||
|
|
"learning_rate": 0.000432,
|
||
|
|
"loss": 6.2858,
|
||
|
|
"mean_token_accuracy": 0.12493444457650185,
|
||
|
|
"num_tokens": 1596140.0,
|
||
|
|
"step": 865
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.3528955459594725,
|
||
|
|
"epoch": 0.048680860588087176,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.0004345,
|
||
|
|
"loss": 6.1884,
|
||
|
|
"mean_token_accuracy": 0.12042548209428787,
|
||
|
|
"num_tokens": 1604811.0,
|
||
|
|
"step": 870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.2718658447265625,
|
||
|
|
"epoch": 0.04896063564893825,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.000437,
|
||
|
|
"loss": 6.1916,
|
||
|
|
"mean_token_accuracy": 0.12246528714895248,
|
||
|
|
"num_tokens": 1613975.0,
|
||
|
|
"step": 875
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.39298005104065,
|
||
|
|
"epoch": 0.04924041070978933,
|
||
|
|
"grad_norm": 1.0859375,
|
||
|
|
"learning_rate": 0.0004395,
|
||
|
|
"loss": 6.2383,
|
||
|
|
"mean_token_accuracy": 0.12414250522851944,
|
||
|
|
"num_tokens": 1622571.0,
|
||
|
|
"step": 880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.373995971679688,
|
||
|
|
"epoch": 0.04952018577064041,
|
||
|
|
"grad_norm": 1.1875,
|
||
|
|
"learning_rate": 0.000442,
|
||
|
|
"loss": 6.3129,
|
||
|
|
"mean_token_accuracy": 0.12346948832273483,
|
||
|
|
"num_tokens": 1631203.0,
|
||
|
|
"step": 885
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.396008825302124,
|
||
|
|
"epoch": 0.049799960831491484,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.0004445,
|
||
|
|
"loss": 6.3239,
|
||
|
|
"mean_token_accuracy": 0.1246532879769802,
|
||
|
|
"num_tokens": 1640170.0,
|
||
|
|
"step": 890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.244958019256591,
|
||
|
|
"epoch": 0.050079735892342554,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.000447,
|
||
|
|
"loss": 6.2958,
|
||
|
|
"mean_token_accuracy": 0.12313124164938927,
|
||
|
|
"num_tokens": 1649039.0,
|
||
|
|
"step": 895
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.412044620513916,
|
||
|
|
"epoch": 0.05035951095319363,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.00044950000000000003,
|
||
|
|
"loss": 6.2599,
|
||
|
|
"mean_token_accuracy": 0.1271291956305504,
|
||
|
|
"num_tokens": 1658445.0,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.195079517364502,
|
||
|
|
"epoch": 0.05063928601404471,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 0.00045200000000000004,
|
||
|
|
"loss": 6.1302,
|
||
|
|
"mean_token_accuracy": 0.13331740349531174,
|
||
|
|
"num_tokens": 1667097.0,
|
||
|
|
"step": 905
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.290958261489868,
|
||
|
|
"epoch": 0.050919061074895786,
|
||
|
|
"grad_norm": 1.234375,
|
||
|
|
"learning_rate": 0.00045450000000000004,
|
||
|
|
"loss": 6.2619,
|
||
|
|
"mean_token_accuracy": 0.11674268618226051,
|
||
|
|
"num_tokens": 1676874.0,
|
||
|
|
"step": 910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.34834885597229,
|
||
|
|
"epoch": 0.05119883613574686,
|
||
|
|
"grad_norm": 1.203125,
|
||
|
|
"learning_rate": 0.00045700000000000005,
|
||
|
|
"loss": 6.2019,
|
||
|
|
"mean_token_accuracy": 0.12399802803993225,
|
||
|
|
"num_tokens": 1685547.0,
|
||
|
|
"step": 915
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.189801597595215,
|
||
|
|
"epoch": 0.05147861119659793,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.00045950000000000006,
|
||
|
|
"loss": 6.191,
|
||
|
|
"mean_token_accuracy": 0.12718559354543685,
|
||
|
|
"num_tokens": 1693753.0,
|
||
|
|
"step": 920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.277320146560669,
|
||
|
|
"epoch": 0.05175838625744901,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.000462,
|
||
|
|
"loss": 6.1807,
|
||
|
|
"mean_token_accuracy": 0.12505132406949998,
|
||
|
|
"num_tokens": 1703747.0,
|
||
|
|
"step": 925
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.1411545753479,
|
||
|
|
"epoch": 0.05203816131830009,
|
||
|
|
"grad_norm": 1.2734375,
|
||
|
|
"learning_rate": 0.0004645,
|
||
|
|
"loss": 6.1278,
|
||
|
|
"mean_token_accuracy": 0.12806382775306702,
|
||
|
|
"num_tokens": 1712656.0,
|
||
|
|
"step": 930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.263706398010254,
|
||
|
|
"epoch": 0.052317936379151164,
|
||
|
|
"grad_norm": 1.2109375,
|
||
|
|
"learning_rate": 0.000467,
|
||
|
|
"loss": 6.2035,
|
||
|
|
"mean_token_accuracy": 0.12317869812250137,
|
||
|
|
"num_tokens": 1722695.0,
|
||
|
|
"step": 935
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.419267845153809,
|
||
|
|
"epoch": 0.05259771144000224,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.0004695,
|
||
|
|
"loss": 6.3361,
|
||
|
|
"mean_token_accuracy": 0.1168564572930336,
|
||
|
|
"num_tokens": 1733003.0,
|
||
|
|
"step": 940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.101446342468262,
|
||
|
|
"epoch": 0.05287748650085331,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.000472,
|
||
|
|
"loss": 6.1556,
|
||
|
|
"mean_token_accuracy": 0.12313281893730163,
|
||
|
|
"num_tokens": 1742393.0,
|
||
|
|
"step": 945
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.349677515029907,
|
||
|
|
"epoch": 0.05315726156170439,
|
||
|
|
"grad_norm": 1.21875,
|
||
|
|
"learning_rate": 0.0004745,
|
||
|
|
"loss": 6.2524,
|
||
|
|
"mean_token_accuracy": 0.12328635305166244,
|
||
|
|
"num_tokens": 1752098.0,
|
||
|
|
"step": 950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.1572033882141115,
|
||
|
|
"epoch": 0.053437036622555466,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 0.000477,
|
||
|
|
"loss": 6.0447,
|
||
|
|
"mean_token_accuracy": 0.12816244289278983,
|
||
|
|
"num_tokens": 1760835.0,
|
||
|
|
"step": 955
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.283704566955566,
|
||
|
|
"epoch": 0.05371681168340654,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.0004795,
|
||
|
|
"loss": 6.2603,
|
||
|
|
"mean_token_accuracy": 0.12562395483255387,
|
||
|
|
"num_tokens": 1770084.0,
|
||
|
|
"step": 960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.192875766754151,
|
||
|
|
"epoch": 0.05399658674425762,
|
||
|
|
"grad_norm": 1.0625,
|
||
|
|
"learning_rate": 0.000482,
|
||
|
|
"loss": 6.2888,
|
||
|
|
"mean_token_accuracy": 0.11796638444066047,
|
||
|
|
"num_tokens": 1779321.0,
|
||
|
|
"step": 965
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.382868099212646,
|
||
|
|
"epoch": 0.05427636180510869,
|
||
|
|
"grad_norm": 1.1015625,
|
||
|
|
"learning_rate": 0.0004845,
|
||
|
|
"loss": 6.2053,
|
||
|
|
"mean_token_accuracy": 0.12535162419080734,
|
||
|
|
"num_tokens": 1789268.0,
|
||
|
|
"step": 970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.16242561340332,
|
||
|
|
"epoch": 0.05455613686595977,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.000487,
|
||
|
|
"loss": 6.1851,
|
||
|
|
"mean_token_accuracy": 0.12277970910072326,
|
||
|
|
"num_tokens": 1797765.0,
|
||
|
|
"step": 975
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.156621074676513,
|
||
|
|
"epoch": 0.054835911926810844,
|
||
|
|
"grad_norm": 1.1171875,
|
||
|
|
"learning_rate": 0.0004895,
|
||
|
|
"loss": 6.1426,
|
||
|
|
"mean_token_accuracy": 0.13391674682497978,
|
||
|
|
"num_tokens": 1807110.0,
|
||
|
|
"step": 980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.318566751480103,
|
||
|
|
"epoch": 0.05511568698766192,
|
||
|
|
"grad_norm": 1.1328125,
|
||
|
|
"learning_rate": 0.000492,
|
||
|
|
"loss": 6.2306,
|
||
|
|
"mean_token_accuracy": 0.1218092642724514,
|
||
|
|
"num_tokens": 1816767.0,
|
||
|
|
"step": 985
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.220181655883789,
|
||
|
|
"epoch": 0.055395462048513,
|
||
|
|
"grad_norm": 1.15625,
|
||
|
|
"learning_rate": 0.0004945,
|
||
|
|
"loss": 6.2397,
|
||
|
|
"mean_token_accuracy": 0.12567930296063423,
|
||
|
|
"num_tokens": 1826012.0,
|
||
|
|
"step": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.257971334457397,
|
||
|
|
"epoch": 0.05567523710936407,
|
||
|
|
"grad_norm": 1.0234375,
|
||
|
|
"learning_rate": 0.000497,
|
||
|
|
"loss": 6.2,
|
||
|
|
"mean_token_accuracy": 0.12385892570018768,
|
||
|
|
"num_tokens": 1835355.0,
|
||
|
|
"step": 995
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 6.210480213165283,
|
||
|
|
"epoch": 0.055955012170215146,
|
||
|
|
"grad_norm": 1.078125,
|
||
|
|
"learning_rate": 0.0004995,
|
||
|
|
"loss": 6.1559,
|
||
|
|
"mean_token_accuracy": 0.13025842383503913,
|
||
|
|
"num_tokens": 1844998.0,
|
||
|
|
"step": 1000
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 4000,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"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": 2700869584896000.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|