Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 444 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.4235. Min Act: -0.1700

Data Index: 552418 (C4 (Web Text))

Max Activating Token Index: 936

Text #1

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.3610. Min Act: -0.1700

Data Index: 232874 (C4 (Web Text))

Max Activating Token Index: 867

Text #2

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.2360. Min Act: -0.1700

Data Index: 755565 (C4 (Web Text))

Max Activating Token Index: 839

Text #3

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.2360. Min Act: -0.1700

Data Index: 621833 (C4 (Web Text))

Max Activating Token Index: 1012

Text #4

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.1735. Min Act: -0.1700

Data Index: 862652 (C4 (Web Text))

Max Activating Token Index: 937

Text #5

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.2047. Min Act: -0.1700

Data Index: 1357960 (C4 (Web Text))

Max Activating Token Index: 853

Text #6

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.1422. Min Act: -0.1699

Data Index: 268658 (C4 (Web Text))

Max Activating Token Index: 975

Text #7

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.1110. Min Act: -0.1698

Data Index: 166470 (C4 (Web Text))

Max Activating Token Index: 442

Text #8

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.1110. Min Act: -0.1700

Data Index: 891077 (C4 (Web Text))

Max Activating Token Index: 726

Text #9

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.1110. Min Act: -0.1700

Data Index: 22222 (C4 (Web Text))

Max Activating Token Index: 621

Text #10

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.0485. Min Act: -0.1686

Data Index: 416080 (C4 (Web Text))

Max Activating Token Index: 683

Text #11

Max Range: 8.4235. Min Range: -8.4235

Max Act: 8.0485. Min Act: -0.1700

Data Index: 1198463 (C4 (Web Text))

Max Activating Token Index: 411

Text #12

Max Range: 8.4235. Min Range: -8.4235

Max Act: 7.9235. Min Act: -0.1699

Data Index: 1141238 (C4 (Web Text))

Max Activating Token Index: 413

Text #13

Max Range: 8.4235. Min Range: -8.4235

Max Act: 7.9547. Min Act: -0.1700

Data Index: 559106 (C4 (Web Text))

Max Activating Token Index: 629

Text #14

Max Range: 8.4235. Min Range: -8.4235

Max Act: 7.9547. Min Act: -0.1700

Data Index: 1122911 (C4 (Web Text))

Max Activating Token Index: 823

Text #15

Max Range: 8.4235. Min Range: -8.4235

Max Act: 7.8922. Min Act: -0.1700

Data Index: 74935 (C4 (Web Text))

Max Activating Token Index: 27

Text #16

Max Range: 8.4235. Min Range: -8.4235

Max Act: 7.8610. Min Act: -0.1700

Data Index: 1313203 (C4 (Web Text))

Max Activating Token Index: 440

Text #17

Max Range: 8.4235. Min Range: -8.4235

Max Act: 7.9235. Min Act: -0.1700

Data Index: 487048 (C4 (Web Text))

Max Activating Token Index: 484

Text #18

Max Range: 8.4235. Min Range: -8.4235

Max Act: 7.8922. Min Act: -0.1700

Data Index: 786263 (C4 (Web Text))

Max Activating Token Index: 371

Text #19

Max Range: 8.4235. Min Range: -8.4235

Max Act: 7.9547. Min Act: -0.1700

Data Index: 739140 (C4 (Web Text))

Max Activating Token Index: 832