Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1547 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0042. Min Act: -0.0003

Data Index: 1113961 (C4 (Web Text))

Max Activating Token Index: 153

Text #1

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0041. Min Act: -0.0003

Data Index: 1393321 (Python Code)

Max Activating Token Index: 532

Text #2

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0041. Min Act: -0.0003

Data Index: 246093 (C4 (Web Text))

Max Activating Token Index: 252

Text #3

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0041. Min Act: -0.0003

Data Index: 1150944 (C4 (Web Text))

Max Activating Token Index: 141

Text #4

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0004

Data Index: 809022 (C4 (Web Text))

Max Activating Token Index: 291

Text #5

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0004

Data Index: 1178313 (C4 (Web Text))

Max Activating Token Index: 990

Text #6

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0003

Data Index: 330550 (C4 (Web Text))

Max Activating Token Index: 584

Text #7

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0003

Data Index: 1604337 (Python Code)

Max Activating Token Index: 86

Text #8

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0003

Data Index: 1366689 (Python Code)

Max Activating Token Index: 717

Text #9

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0004

Data Index: 389300 (C4 (Web Text))

Max Activating Token Index: 149

Text #10

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0003

Data Index: 643935 (C4 (Web Text))

Max Activating Token Index: 339

Text #11

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0003

Data Index: 1546993 (Python Code)

Max Activating Token Index: 604

Text #12

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0038. Min Act: -0.0004

Data Index: 1184320 (C4 (Web Text))

Max Activating Token Index: 150

Text #13

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0039. Min Act: -0.0003

Data Index: 487474 (C4 (Web Text))

Max Activating Token Index: 169

Text #14

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0038. Min Act: -0.0003

Data Index: 881611 (C4 (Web Text))

Max Activating Token Index: 159

Text #15

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0038. Min Act: -0.0003

Data Index: 1154821 (C4 (Web Text))

Max Activating Token Index: 153

Text #16

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0038. Min Act: -0.0004

Data Index: 1185480 (C4 (Web Text))

Max Activating Token Index: 24

Text #17

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0038. Min Act: -0.0003

Data Index: 1640900 (Python Code)

Max Activating Token Index: 969

Text #18

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0038. Min Act: -0.0003

Data Index: 930837 (C4 (Web Text))

Max Activating Token Index: 51

Text #19

Max Range: 0.0042. Min Range: -0.0042

Max Act: 0.0037. Min Act: -0.0003

Data Index: 145446 (C4 (Web Text))

Max Activating Token Index: 262