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

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

Neuron 1493 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.4254. Min Act: -0.1700

Data Index: 305329 (C4 (Web Text))

Max Activating Token Index: 253

Text #1

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.3313. Min Act: -0.1700

Data Index: 100806 (C4 (Web Text))

Max Activating Token Index: 328

Text #2

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.2999. Min Act: -0.1700

Data Index: 177731 (C4 (Web Text))

Max Activating Token Index: 334

Text #3

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.2999. Min Act: -0.1700

Data Index: 983779 (C4 (Web Text))

Max Activating Token Index: 483

Text #4

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.2370. Min Act: -0.1700

Data Index: 608350 (C4 (Web Text))

Max Activating Token Index: 284

Text #5

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.2370. Min Act: -0.1700

Data Index: 1095266 (C4 (Web Text))

Max Activating Token Index: 811

Text #6

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.1898. Min Act: -0.1700

Data Index: 271538 (C4 (Web Text))

Max Activating Token Index: 738

Text #7

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.1898. Min Act: -0.1700

Data Index: 1206701 (C4 (Web Text))

Max Activating Token Index: 46

Text #8

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.1741. Min Act: -0.1700

Data Index: 907643 (C4 (Web Text))

Max Activating Token Index: 359

Text #9

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.1583. Min Act: -0.1700

Data Index: 1189733 (C4 (Web Text))

Max Activating Token Index: 379

Text #10

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.1898. Min Act: -0.1700

Data Index: 415146 (C4 (Web Text))

Max Activating Token Index: 506

Text #11

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.1426. Min Act: -0.1700

Data Index: 1270313 (C4 (Web Text))

Max Activating Token Index: 311

Text #12

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.1111. Min Act: -0.1700

Data Index: 1322449 (C4 (Web Text))

Max Activating Token Index: 614

Text #13

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.0953. Min Act: -0.1700

Data Index: 795357 (C4 (Web Text))

Max Activating Token Index: 338

Text #14

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.0953. Min Act: -0.1700

Data Index: 782316 (C4 (Web Text))

Max Activating Token Index: 598

Text #15

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.1111. Min Act: -0.1700

Data Index: 1416324 (Python Code)

Max Activating Token Index: 607

Text #16

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.0953. Min Act: -0.1700

Data Index: 1016408 (C4 (Web Text))

Max Activating Token Index: 157

Text #17

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.0795. Min Act: -0.1700

Data Index: 301636 (C4 (Web Text))

Max Activating Token Index: 774

Text #18

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.0795. Min Act: -0.1700

Data Index: 957928 (C4 (Web Text))

Max Activating Token Index: 60

Text #19

Max Range: 3.4254. Min Range: -3.4254

Max Act: 3.0795. Min Act: -0.1700

Data Index: 974986 (C4 (Web Text))

Max Activating Token Index: 1013