Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 24 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 4.7091. Min Range: -4.7091

Max Act: 4.7091. Min Act: -0.0005

Data Index: 544368 (C4 (Web Text))

Max Activating Token Index: 17

Text #1

Max Range: 4.7091. Min Range: -4.7091

Max Act: 4.3547. Min Act: -0.0005

Data Index: 395027 (C4 (Web Text))

Max Activating Token Index: 18

Text #2

Max Range: 4.7091. Min Range: -4.7091

Max Act: 4.0753. Min Act: -0.0005

Data Index: 286213 (C4 (Web Text))

Max Activating Token Index: 11

Text #3

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.9488. Min Act: -0.0005

Data Index: 236138 (C4 (Web Text))

Max Activating Token Index: 815

Text #4

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.8899. Min Act: -0.0005

Data Index: 669699 (C4 (Web Text))

Max Activating Token Index: 22

Text #5

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.5906. Min Act: -0.0005

Data Index: 1243178 (C4 (Web Text))

Max Activating Token Index: 932

Text #6

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.7348. Min Act: -0.0005

Data Index: 839306 (C4 (Web Text))

Max Activating Token Index: 108

Text #7

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.8632. Min Act: -0.0005

Data Index: 144275 (C4 (Web Text))

Max Activating Token Index: 710

Text #8

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.9546. Min Act: -0.0004

Data Index: 1246371 (C4 (Web Text))

Max Activating Token Index: 65

Text #9

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.5043. Min Act: -0.0005

Data Index: 1100221 (C4 (Web Text))

Max Activating Token Index: 341

Text #10

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.7414. Min Act: -0.0005

Data Index: 518516 (C4 (Web Text))

Max Activating Token Index: 140

Text #11

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.5875. Min Act: -0.0005

Data Index: 271590 (C4 (Web Text))

Max Activating Token Index: 50

Text #12

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.4291. Min Act: -0.0005

Data Index: 395170 (C4 (Web Text))

Max Activating Token Index: 45

Text #13

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.3849. Min Act: -0.0005

Data Index: 15402 (C4 (Web Text))

Max Activating Token Index: 24

Text #14

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.1996. Min Act: -0.0005

Data Index: 329548 (C4 (Web Text))

Max Activating Token Index: 317

Text #15

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.3055. Min Act: -0.0005

Data Index: 423777 (C4 (Web Text))

Max Activating Token Index: 64

Text #16

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.4216. Min Act: -0.0005

Data Index: 925501 (C4 (Web Text))

Max Activating Token Index: 43

Text #17

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.1097. Min Act: -0.0005

Data Index: 463509 (C4 (Web Text))

Max Activating Token Index: 830

Text #18

Max Range: 4.7091. Min Range: -4.7091

Max Act: 3.0943. Min Act: -0.0005

Data Index: 575140 (C4 (Web Text))

Max Activating Token Index: 320

Text #19

Max Range: 4.7091. Min Range: -4.7091

Max Act: 2.9712. Min Act: -0.0005

Data Index: 408520 (C4 (Web Text))

Max Activating Token Index: 966