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

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

Neuron 1647 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: 5.8333. Min Range: -5.8333

Max Act: 5.8333. Min Act: -0.1700

Data Index: 512758 (C4 (Web Text))

Max Activating Token Index: 486

Text #1

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.7395. Min Act: -0.1700

Data Index: 5947 (C4 (Web Text))

Max Activating Token Index: 983

Text #2

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.4270. Min Act: -0.1700

Data Index: 142337 (C4 (Web Text))

Max Activating Token Index: 322

Text #3

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.4583. Min Act: -0.1700

Data Index: 1164986 (C4 (Web Text))

Max Activating Token Index: 953

Text #4

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.4270. Min Act: -0.1700

Data Index: 1303838 (C4 (Web Text))

Max Activating Token Index: 100

Text #5

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3333. Min Act: -0.1700

Data Index: 257327 (C4 (Web Text))

Max Activating Token Index: 406

Text #6

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3333. Min Act: -0.1700

Data Index: 96994 (C4 (Web Text))

Max Activating Token Index: 979

Text #7

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3958. Min Act: -0.1700

Data Index: 947978 (C4 (Web Text))

Max Activating Token Index: 868

Text #8

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3958. Min Act: -0.1700

Data Index: 1080008 (C4 (Web Text))

Max Activating Token Index: 1013

Text #9

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.4270. Min Act: -0.1700

Data Index: 1124975 (C4 (Web Text))

Max Activating Token Index: 988

Text #10

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3645. Min Act: -0.1700

Data Index: 663944 (C4 (Web Text))

Max Activating Token Index: 280

Text #11

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3645. Min Act: -0.1700

Data Index: 1177429 (C4 (Web Text))

Max Activating Token Index: 253

Text #12

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3645. Min Act: -0.1700

Data Index: 47855 (C4 (Web Text))

Max Activating Token Index: 380

Text #13

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3333. Min Act: -0.1700

Data Index: 1228962 (C4 (Web Text))

Max Activating Token Index: 755

Text #14

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3333. Min Act: -0.1700

Data Index: 522919 (C4 (Web Text))

Max Activating Token Index: 373

Text #15

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3333. Min Act: -0.1700

Data Index: 662700 (C4 (Web Text))

Max Activating Token Index: 250

Text #16

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3333. Min Act: -0.1700

Data Index: 98510 (C4 (Web Text))

Max Activating Token Index: 642

Text #17

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3020. Min Act: -0.1700

Data Index: 931975 (C4 (Web Text))

Max Activating Token Index: 685

Text #18

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.3020. Min Act: -0.1700

Data Index: 347377 (C4 (Web Text))

Max Activating Token Index: 704

Text #19

Max Range: 5.8333. Min Range: -5.8333

Max Act: 5.2708. Min Act: -0.1700

Data Index: 10839 (C4 (Web Text))

Max Activating Token Index: 88