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

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

Neuron 1892 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: 5.4983. Min Range: -5.4983

Max Act: 5.4983. Min Act: -0.1700

Data Index: 1097221 (C4 (Web Text))

Max Activating Token Index: 15

Text #1

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.4670. Min Act: -0.1700

Data Index: 324020 (C4 (Web Text))

Max Activating Token Index: 19

Text #2

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.3420. Min Act: -0.1700

Data Index: 225517 (C4 (Web Text))

Max Activating Token Index: 1008

Text #3

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.4045. Min Act: -0.1700

Data Index: 1214863 (C4 (Web Text))

Max Activating Token Index: 922

Text #4

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.4045. Min Act: -0.1700

Data Index: 659414 (C4 (Web Text))

Max Activating Token Index: 495

Text #5

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.2170. Min Act: -0.1700

Data Index: 919428 (C4 (Web Text))

Max Activating Token Index: 976

Text #6

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.2170. Min Act: -0.1700

Data Index: 70618 (C4 (Web Text))

Max Activating Token Index: 469

Text #7

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.3108. Min Act: -0.1700

Data Index: 23179 (C4 (Web Text))

Max Activating Token Index: 271

Text #8

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.3108. Min Act: -0.1700

Data Index: 484141 (C4 (Web Text))

Max Activating Token Index: 503

Text #9

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.3108. Min Act: -0.1700

Data Index: 905525 (C4 (Web Text))

Max Activating Token Index: 258

Text #10

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.0608. Min Act: -0.1700

Data Index: 1082174 (C4 (Web Text))

Max Activating Token Index: 143

Text #11

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.1233. Min Act: -0.1700

Data Index: 17578 (C4 (Web Text))

Max Activating Token Index: 573

Text #12

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.0295. Min Act: -0.1700

Data Index: 1336091 (C4 (Web Text))

Max Activating Token Index: 253

Text #13

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.0608. Min Act: -0.1700

Data Index: 403720 (C4 (Web Text))

Max Activating Token Index: 810

Text #14

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.0608. Min Act: -0.1700

Data Index: 709703 (C4 (Web Text))

Max Activating Token Index: 866

Text #15

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.1233. Min Act: -0.1700

Data Index: 2158 (C4 (Web Text))

Max Activating Token Index: 481

Text #16

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.0608. Min Act: -0.1700

Data Index: 1073222 (C4 (Web Text))

Max Activating Token Index: 844

Text #17

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.1233. Min Act: -0.1700

Data Index: 1170172 (C4 (Web Text))

Max Activating Token Index: 170

Text #18

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.0920. Min Act: -0.1700

Data Index: 351513 (C4 (Web Text))

Max Activating Token Index: 67

Text #19

Max Range: 5.4983. Min Range: -5.4983

Max Act: 5.0608. Min Act: -0.1700

Data Index: 175251 (C4 (Web Text))

Max Activating Token Index: 12