Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2980 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.4601. Min Act: -0.0001

Data Index: 848669 (C4 (Web Text))

Max Activating Token Index: 7

Text #1

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.3571. Min Act: -0.0001

Data Index: 81159 (C4 (Web Text))

Max Activating Token Index: 618

Text #2

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.3221. Min Act: -0.0001

Data Index: 127220 (C4 (Web Text))

Max Activating Token Index: 638

Text #3

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.2674. Min Act: -0.0001

Data Index: 929913 (C4 (Web Text))

Max Activating Token Index: 548

Text #4

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.2259. Min Act: -0.0001

Data Index: 915038 (C4 (Web Text))

Max Activating Token Index: 1017

Text #5

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1990. Min Act: -0.0001

Data Index: 1207865 (C4 (Web Text))

Max Activating Token Index: 939

Text #6

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1858. Min Act: -0.0001

Data Index: 770113 (C4 (Web Text))

Max Activating Token Index: 214

Text #7

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1614. Min Act: -0.0001

Data Index: 498812 (C4 (Web Text))

Max Activating Token Index: 486

Text #8

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1587. Min Act: -0.0001

Data Index: 94190 (C4 (Web Text))

Max Activating Token Index: 12

Text #9

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1582. Min Act: -0.0001

Data Index: 203840 (C4 (Web Text))

Max Activating Token Index: 274

Text #10

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1112. Min Act: -0.0001

Data Index: 981280 (C4 (Web Text))

Max Activating Token Index: 783

Text #11

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1062. Min Act: -0.0001

Data Index: 211743 (C4 (Web Text))

Max Activating Token Index: 80

Text #12

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1866. Min Act: -0.0001

Data Index: 13814 (C4 (Web Text))

Max Activating Token Index: 328

Text #13

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.0673. Min Act: -0.0001

Data Index: 1334207 (C4 (Web Text))

Max Activating Token Index: 728

Text #14

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.1173. Min Act: -0.0001

Data Index: 1208999 (C4 (Web Text))

Max Activating Token Index: 502

Text #15

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.0833. Min Act: -0.0001

Data Index: 900430 (C4 (Web Text))

Max Activating Token Index: 734

Text #16

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.0145. Min Act: -0.0001

Data Index: 10011 (C4 (Web Text))

Max Activating Token Index: 306

Text #17

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.0693. Min Act: -0.0001

Data Index: 1483209 (Python Code)

Max Activating Token Index: 29

Text #18

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.0987. Min Act: -0.0001

Data Index: 281005 (C4 (Web Text))

Max Activating Token Index: 790

Text #19

Max Range: 1.4601. Min Range: -1.4601

Max Act: 1.0597. Min Act: -0.0001

Data Index: 340458 (C4 (Web Text))

Max Activating Token Index: 21