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

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

Neuron 90 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: 7.7960. Min Range: -7.7960

Max Act: 7.7960. Min Act: -0.1700

Data Index: 1350039 (C4 (Web Text))

Max Activating Token Index: 311

Text #1

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.6710. Min Act: -0.1699

Data Index: 320458 (C4 (Web Text))

Max Activating Token Index: 432

Text #2

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.6085. Min Act: -0.1700

Data Index: 655023 (C4 (Web Text))

Max Activating Token Index: 940

Text #3

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.4523. Min Act: -0.1700

Data Index: 434567 (C4 (Web Text))

Max Activating Token Index: 246

Text #4

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.5148. Min Act: -0.1700

Data Index: 1349215 (C4 (Web Text))

Max Activating Token Index: 1003

Text #5

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.5148. Min Act: -0.1700

Data Index: 329142 (C4 (Web Text))

Max Activating Token Index: 382

Text #6

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.3273. Min Act: -0.1700

Data Index: 1347630 (C4 (Web Text))

Max Activating Token Index: 177

Text #7

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.4210. Min Act: -0.1700

Data Index: 193260 (C4 (Web Text))

Max Activating Token Index: 254

Text #8

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.3898. Min Act: -0.1700

Data Index: 1237722 (C4 (Web Text))

Max Activating Token Index: 948

Text #9

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.3585. Min Act: -0.1700

Data Index: 577834 (C4 (Web Text))

Max Activating Token Index: 255

Text #10

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.2960. Min Act: -0.1700

Data Index: 1282635 (C4 (Web Text))

Max Activating Token Index: 756

Text #11

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.2960. Min Act: -0.1700

Data Index: 313290 (C4 (Web Text))

Max Activating Token Index: 234

Text #12

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.3585. Min Act: -0.1700

Data Index: 916543 (C4 (Web Text))

Max Activating Token Index: 846

Text #13

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.4210. Min Act: -0.1700

Data Index: 1206179 (C4 (Web Text))

Max Activating Token Index: 514

Text #14

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.3585. Min Act: -0.1700

Data Index: 1189066 (C4 (Web Text))

Max Activating Token Index: 995

Text #15

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.3273. Min Act: -0.1700

Data Index: 728539 (C4 (Web Text))

Max Activating Token Index: 226

Text #16

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.2960. Min Act: -0.1700

Data Index: 91623 (C4 (Web Text))

Max Activating Token Index: 919

Text #17

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.2648. Min Act: -0.1700

Data Index: 1050276 (C4 (Web Text))

Max Activating Token Index: 898

Text #18

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.2335. Min Act: -0.1700

Data Index: 1288802 (C4 (Web Text))

Max Activating Token Index: 429

Text #19

Max Range: 7.7960. Min Range: -7.7960

Max Act: 7.2960. Min Act: -0.1700

Data Index: 1070198 (C4 (Web Text))

Max Activating Token Index: 643