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

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

Neuron 1052 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: 5.1777. Min Range: -5.1777

Max Act: 5.1777. Min Act: -0.1700

Data Index: 1108025 (C4 (Web Text))

Max Activating Token Index: 95

Text #1

Max Range: 5.1777. Min Range: -5.1777

Max Act: 5.0839. Min Act: -0.1700

Data Index: 427705 (C4 (Web Text))

Max Activating Token Index: 226

Text #2

Max Range: 5.1777. Min Range: -5.1777

Max Act: 5.0527. Min Act: -0.1698

Data Index: 814374 (C4 (Web Text))

Max Activating Token Index: 120

Text #3

Max Range: 5.1777. Min Range: -5.1777

Max Act: 5.0527. Min Act: -0.1699

Data Index: 713102 (C4 (Web Text))

Max Activating Token Index: 196

Text #4

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.9277. Min Act: -0.1700

Data Index: 706202 (C4 (Web Text))

Max Activating Token Index: 103

Text #5

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8339. Min Act: -0.1700

Data Index: 850251 (C4 (Web Text))

Max Activating Token Index: 193

Text #6

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8964. Min Act: -0.1700

Data Index: 1244031 (C4 (Web Text))

Max Activating Token Index: 69

Text #7

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8652. Min Act: -0.1700

Data Index: 1297703 (C4 (Web Text))

Max Activating Token Index: 930

Text #8

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8339. Min Act: -0.1700

Data Index: 276657 (C4 (Web Text))

Max Activating Token Index: 73

Text #9

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8652. Min Act: -0.1700

Data Index: 367801 (C4 (Web Text))

Max Activating Token Index: 24

Text #10

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8339. Min Act: -0.1700

Data Index: 363369 (C4 (Web Text))

Max Activating Token Index: 176

Text #11

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8027. Min Act: -0.1700

Data Index: 1292081 (C4 (Web Text))

Max Activating Token Index: 853

Text #12

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8339. Min Act: -0.1699

Data Index: 56529 (C4 (Web Text))

Max Activating Token Index: 93

Text #13

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8964. Min Act: -0.1700

Data Index: 1046866 (C4 (Web Text))

Max Activating Token Index: 404

Text #14

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8339. Min Act: -0.1700

Data Index: 394942 (C4 (Web Text))

Max Activating Token Index: 974

Text #15

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.7714. Min Act: -0.1700

Data Index: 400732 (C4 (Web Text))

Max Activating Token Index: 22

Text #16

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8027. Min Act: -0.1700

Data Index: 1045765 (C4 (Web Text))

Max Activating Token Index: 39

Text #17

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.8027. Min Act: -0.1700

Data Index: 3731 (C4 (Web Text))

Max Activating Token Index: 652

Text #18

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.7558. Min Act: -0.1700

Data Index: 1169845 (C4 (Web Text))

Max Activating Token Index: 230

Text #19

Max Range: 5.1777. Min Range: -5.1777

Max Act: 4.7246. Min Act: -0.1700

Data Index: 448467 (C4 (Web Text))

Max Activating Token Index: 85