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

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

Neuron 950 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: 5.3217. Min Range: -5.3217

Max Act: 5.3217. Min Act: -0.1700

Data Index: 330859 (C4 (Web Text))

Max Activating Token Index: 440

Text #1

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.2592. Min Act: -0.1700

Data Index: 567311 (C4 (Web Text))

Max Activating Token Index: 289

Text #2

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1967. Min Act: -0.1700

Data Index: 1142963 (C4 (Web Text))

Max Activating Token Index: 690

Text #3

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1654. Min Act: -0.1700

Data Index: 354261 (C4 (Web Text))

Max Activating Token Index: 412

Text #4

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1342. Min Act: -0.1700

Data Index: 1104353 (C4 (Web Text))

Max Activating Token Index: 340

Text #5

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1342. Min Act: -0.1700

Data Index: 42077 (C4 (Web Text))

Max Activating Token Index: 161

Text #6

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1342. Min Act: -0.1700

Data Index: 77159 (C4 (Web Text))

Max Activating Token Index: 474

Text #7

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1342. Min Act: -0.1700

Data Index: 1163604 (C4 (Web Text))

Max Activating Token Index: 91

Text #8

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1654. Min Act: -0.1700

Data Index: 669366 (C4 (Web Text))

Max Activating Token Index: 299

Text #9

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1654. Min Act: -0.1700

Data Index: 908562 (C4 (Web Text))

Max Activating Token Index: 42

Text #10

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1654. Min Act: -0.1700

Data Index: 1034893 (C4 (Web Text))

Max Activating Token Index: 624

Text #11

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1654. Min Act: -0.1700

Data Index: 208999 (C4 (Web Text))

Max Activating Token Index: 260

Text #12

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1654. Min Act: -0.1700

Data Index: 263225 (C4 (Web Text))

Max Activating Token Index: 47

Text #13

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1029. Min Act: -0.1700

Data Index: 56226 (C4 (Web Text))

Max Activating Token Index: 442

Text #14

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1029. Min Act: -0.1700

Data Index: 696836 (C4 (Web Text))

Max Activating Token Index: 404

Text #15

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.0717. Min Act: -0.1700

Data Index: 1352893 (C4 (Web Text))

Max Activating Token Index: 443

Text #16

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.0404. Min Act: -0.1700

Data Index: 597722 (C4 (Web Text))

Max Activating Token Index: 340

Text #17

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1029. Min Act: -0.1700

Data Index: 632598 (C4 (Web Text))

Max Activating Token Index: 61

Text #18

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.0717. Min Act: -0.1700

Data Index: 830713 (C4 (Web Text))

Max Activating Token Index: 689

Text #19

Max Range: 5.3217. Min Range: -5.3217

Max Act: 5.1029. Min Act: -0.1700

Data Index: 548479 (C4 (Web Text))

Max Activating Token Index: 177