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

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

Neuron 1056 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: 3.5899. Min Range: -3.5899

Max Act: 3.5899. Min Act: -0.1700

Data Index: 706084 (C4 (Web Text))

Max Activating Token Index: 195

Text #1

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.5899. Min Act: -0.1700

Data Index: 514014 (C4 (Web Text))

Max Activating Token Index: 213

Text #2

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.5585. Min Act: -0.1700

Data Index: 122036 (C4 (Web Text))

Max Activating Token Index: 100

Text #3

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.5429. Min Act: -0.1700

Data Index: 841380 (C4 (Web Text))

Max Activating Token Index: 104

Text #4

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.5272. Min Act: -0.1700

Data Index: 377592 (C4 (Web Text))

Max Activating Token Index: 142

Text #5

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.5742. Min Act: -0.1700

Data Index: 367805 (C4 (Web Text))

Max Activating Token Index: 64

Text #6

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.5429. Min Act: -0.1700

Data Index: 936770 (C4 (Web Text))

Max Activating Token Index: 306

Text #7

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.5272. Min Act: -0.1700

Data Index: 97220 (C4 (Web Text))

Max Activating Token Index: 11

Text #8

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4959. Min Act: -0.1700

Data Index: 123910 (C4 (Web Text))

Max Activating Token Index: 50

Text #9

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4959. Min Act: -0.1700

Data Index: 1271082 (C4 (Web Text))

Max Activating Token Index: 557

Text #10

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4959. Min Act: -0.1700

Data Index: 618951 (C4 (Web Text))

Max Activating Token Index: 797

Text #11

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4959. Min Act: -0.1700

Data Index: 735938 (C4 (Web Text))

Max Activating Token Index: 569

Text #12

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4645. Min Act: -0.1700

Data Index: 1009729 (C4 (Web Text))

Max Activating Token Index: 17

Text #13

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4645. Min Act: -0.1700

Data Index: 206313 (C4 (Web Text))

Max Activating Token Index: 4

Text #14

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4802. Min Act: -0.1700

Data Index: 262427 (C4 (Web Text))

Max Activating Token Index: 108

Text #15

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4645. Min Act: -0.1700

Data Index: 1010387 (C4 (Web Text))

Max Activating Token Index: 508

Text #16

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4645. Min Act: -0.1700

Data Index: 815694 (C4 (Web Text))

Max Activating Token Index: 170

Text #17

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4645. Min Act: -0.1700

Data Index: 284354 (C4 (Web Text))

Max Activating Token Index: 174

Text #18

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4645. Min Act: -0.1700

Data Index: 1333262 (C4 (Web Text))

Max Activating Token Index: 516

Text #19

Max Range: 3.5899. Min Range: -3.5899

Max Act: 3.4332. Min Act: -0.1700

Data Index: 951768 (C4 (Web Text))

Max Activating Token Index: 688