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

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

Neuron 2500 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.9135. Min Act: -0.0001

Data Index: 138449 (C4 (Web Text))

Max Activating Token Index: 688

Text #1

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.7237. Min Act: -0.0001

Data Index: 9235 (C4 (Web Text))

Max Activating Token Index: 173

Text #2

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.6880. Min Act: -0.0001

Data Index: 604328 (C4 (Web Text))

Max Activating Token Index: 621

Text #3

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.8913. Min Act: -0.0001

Data Index: 98805 (C4 (Web Text))

Max Activating Token Index: 248

Text #4

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.4105. Min Act: -0.0001

Data Index: 645254 (C4 (Web Text))

Max Activating Token Index: 820

Text #5

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.5683. Min Act: -0.0001

Data Index: 1130550 (C4 (Web Text))

Max Activating Token Index: 369

Text #6

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.3516. Min Act: -0.0001

Data Index: 934093 (C4 (Web Text))

Max Activating Token Index: 565

Text #7

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.2897. Min Act: -0.0001

Data Index: 81377 (C4 (Web Text))

Max Activating Token Index: 858

Text #8

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.1503. Min Act: -0.0001

Data Index: 1139378 (C4 (Web Text))

Max Activating Token Index: 520

Text #9

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.0964. Min Act: -0.0001

Data Index: 1174182 (C4 (Web Text))

Max Activating Token Index: 10

Text #10

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.0558. Min Act: -0.0001

Data Index: 395548 (C4 (Web Text))

Max Activating Token Index: 903

Text #11

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.1481. Min Act: -0.0001

Data Index: 794430 (C4 (Web Text))

Max Activating Token Index: 577

Text #12

Max Range: 3.9135. Min Range: -3.9135

Max Act: 2.9896. Min Act: -0.0001

Data Index: 298026 (C4 (Web Text))

Max Activating Token Index: 1012

Text #13

Max Range: 3.9135. Min Range: -3.9135

Max Act: 2.9888. Min Act: -0.0001

Data Index: 510399 (C4 (Web Text))

Max Activating Token Index: 969

Text #14

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.1195. Min Act: -0.0001

Data Index: 375502 (C4 (Web Text))

Max Activating Token Index: 437

Text #15

Max Range: 3.9135. Min Range: -3.9135

Max Act: 2.9408. Min Act: -0.0001

Data Index: 819957 (C4 (Web Text))

Max Activating Token Index: 238

Text #16

Max Range: 3.9135. Min Range: -3.9135

Max Act: 3.2102. Min Act: -0.0001

Data Index: 1133767 (C4 (Web Text))

Max Activating Token Index: 713

Text #17

Max Range: 3.9135. Min Range: -3.9135

Max Act: 2.9209. Min Act: -0.0001

Data Index: 999989 (C4 (Web Text))

Max Activating Token Index: 811

Text #18

Max Range: 3.9135. Min Range: -3.9135

Max Act: 2.7478. Min Act: -0.0001

Data Index: 656408 (C4 (Web Text))

Max Activating Token Index: 272

Text #19

Max Range: 3.9135. Min Range: -3.9135

Max Act: 2.9422. Min Act: -0.0001

Data Index: 291295 (C4 (Web Text))

Max Activating Token Index: 328