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

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

Neuron 80 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.3247. Min Range: -5.3247

Max Act: 5.3247. Min Act: -0.1700

Data Index: 945755 (C4 (Web Text))

Max Activating Token Index: 696

Text #1

Max Range: 5.3247. Min Range: -5.3247

Max Act: 5.2934. Min Act: -0.1700

Data Index: 462499 (C4 (Web Text))

Max Activating Token Index: 601

Text #2

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.8872. Min Act: -0.1700

Data Index: 351780 (C4 (Web Text))

Max Activating Token Index: 925

Text #3

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.8559. Min Act: -0.1700

Data Index: 896610 (C4 (Web Text))

Max Activating Token Index: 895

Text #4

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6059. Min Act: -0.1700

Data Index: 1174524 (C4 (Web Text))

Max Activating Token Index: 802

Text #5

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.7309. Min Act: -0.1700

Data Index: 33569 (C4 (Web Text))

Max Activating Token Index: 841

Text #6

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.7622. Min Act: -0.1700

Data Index: 829280 (C4 (Web Text))

Max Activating Token Index: 938

Text #7

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6997. Min Act: -0.1700

Data Index: 427180 (C4 (Web Text))

Max Activating Token Index: 220

Text #8

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6997. Min Act: -0.1700

Data Index: 433850 (C4 (Web Text))

Max Activating Token Index: 302

Text #9

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.7309. Min Act: -0.1700

Data Index: 958161 (C4 (Web Text))

Max Activating Token Index: 982

Text #10

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6997. Min Act: -0.1700

Data Index: 538786 (C4 (Web Text))

Max Activating Token Index: 622

Text #11

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6684. Min Act: -0.1700

Data Index: 1064341 (C4 (Web Text))

Max Activating Token Index: 519

Text #12

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6684. Min Act: -0.1700

Data Index: 508007 (C4 (Web Text))

Max Activating Token Index: 931

Text #13

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6684. Min Act: -0.1700

Data Index: 231069 (C4 (Web Text))

Max Activating Token Index: 778

Text #14

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6372. Min Act: -0.1700

Data Index: 18025 (C4 (Web Text))

Max Activating Token Index: 284

Text #15

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6059. Min Act: -0.1700

Data Index: 1067811 (C4 (Web Text))

Max Activating Token Index: 521

Text #16

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.5747. Min Act: -0.1700

Data Index: 610971 (C4 (Web Text))

Max Activating Token Index: 447

Text #17

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6684. Min Act: -0.1700

Data Index: 171202 (C4 (Web Text))

Max Activating Token Index: 984

Text #18

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6372. Min Act: -0.1700

Data Index: 1287311 (C4 (Web Text))

Max Activating Token Index: 593

Text #19

Max Range: 5.3247. Min Range: -5.3247

Max Act: 4.6372. Min Act: -0.1700

Data Index: 605456 (C4 (Web Text))

Max Activating Token Index: 567