Model: GELU Model: 3 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 1029 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.5701. Min Act: -0.1700

Data Index: 291749 (C4 (Web Text))

Max Activating Token Index: 362

Text #1

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.5231. Min Act: -0.1700

Data Index: 544152 (C4 (Web Text))

Max Activating Token Index: 519

Text #2

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.4761. Min Act: -0.1700

Data Index: 474967 (C4 (Web Text))

Max Activating Token Index: 493

Text #3

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.4761. Min Act: -0.1700

Data Index: 643499 (C4 (Web Text))

Max Activating Token Index: 700

Text #4

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.4447. Min Act: -0.1700

Data Index: 1222556 (C4 (Web Text))

Max Activating Token Index: 428

Text #5

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.3977. Min Act: -0.1700

Data Index: 1064916 (C4 (Web Text))

Max Activating Token Index: 149

Text #6

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.3663. Min Act: -0.1700

Data Index: 1062644 (C4 (Web Text))

Max Activating Token Index: 820

Text #7

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.3506. Min Act: -0.1700

Data Index: 755466 (C4 (Web Text))

Max Activating Token Index: 509

Text #8

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.3192. Min Act: -0.1700

Data Index: 1000895 (C4 (Web Text))

Max Activating Token Index: 336

Text #9

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.3349. Min Act: -0.1700

Data Index: 421952 (C4 (Web Text))

Max Activating Token Index: 472

Text #10

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.3349. Min Act: -0.1700

Data Index: 841142 (C4 (Web Text))

Max Activating Token Index: 458

Text #11

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.2878. Min Act: -0.1700

Data Index: 860199 (C4 (Web Text))

Max Activating Token Index: 554

Text #12

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.2407. Min Act: -0.1700

Data Index: 441264 (C4 (Web Text))

Max Activating Token Index: 371

Text #13

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.2721. Min Act: -0.1700

Data Index: 460496 (C4 (Web Text))

Max Activating Token Index: 180

Text #14

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.2564. Min Act: -0.1700

Data Index: 999304 (C4 (Web Text))

Max Activating Token Index: 729

Text #15

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.2564. Min Act: -0.1700

Data Index: 499255 (C4 (Web Text))

Max Activating Token Index: 226

Text #16

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.2092. Min Act: -0.1700

Data Index: 682271 (C4 (Web Text))

Max Activating Token Index: 91

Text #17

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.1935. Min Act: -0.1700

Data Index: 885813 (C4 (Web Text))

Max Activating Token Index: 555

Text #18

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.2092. Min Act: -0.1700

Data Index: 451498 (C4 (Web Text))

Max Activating Token Index: 669

Text #19

Max Range: 3.5701. Min Range: -3.5701

Max Act: 3.2249. Min Act: -0.1700

Data Index: 1103588 (C4 (Web Text))

Max Activating Token Index: 805