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

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

Neuron 2112 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3186. Min Range: -1.3186

Max Act: 1.3186. Min Act: -0.0001

Data Index: 801587 (C4 (Web Text))

Max Activating Token Index: 119

Text #1

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.9152. Min Act: -0.0001

Data Index: 46875 (C4 (Web Text))

Max Activating Token Index: 188

Text #2

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.9088. Min Act: -0.0001

Data Index: 338252 (C4 (Web Text))

Max Activating Token Index: 493

Text #3

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8244. Min Act: -0.0001

Data Index: 395347 (C4 (Web Text))

Max Activating Token Index: 363

Text #4

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8675. Min Act: -0.0000

Data Index: 980096 (C4 (Web Text))

Max Activating Token Index: 134

Text #5

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8593. Min Act: -0.0001

Data Index: 1506117 (Python Code)

Max Activating Token Index: 238

Text #6

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8940. Min Act: -0.0001

Data Index: 387740 (C4 (Web Text))

Max Activating Token Index: 287

Text #7

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8443. Min Act: -0.0001

Data Index: 770301 (C4 (Web Text))

Max Activating Token Index: 50

Text #8

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8422. Min Act: -0.0001

Data Index: 28490 (C4 (Web Text))

Max Activating Token Index: 32

Text #9

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8102. Min Act: -0.0001

Data Index: 1358667 (C4 (Web Text))

Max Activating Token Index: 36

Text #10

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8086. Min Act: -0.0000

Data Index: 888308 (C4 (Web Text))

Max Activating Token Index: 96

Text #11

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8319. Min Act: -0.0001

Data Index: 67374 (C4 (Web Text))

Max Activating Token Index: 250

Text #12

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8342. Min Act: -0.0001

Data Index: 569500 (C4 (Web Text))

Max Activating Token Index: 145

Text #13

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.7572. Min Act: -0.0001

Data Index: 1074704 (C4 (Web Text))

Max Activating Token Index: 71

Text #14

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8020. Min Act: -0.0000

Data Index: 191172 (C4 (Web Text))

Max Activating Token Index: 34

Text #15

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8279. Min Act: -0.0001

Data Index: 982418 (C4 (Web Text))

Max Activating Token Index: 78

Text #16

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8150. Min Act: -0.0001

Data Index: 184999 (C4 (Web Text))

Max Activating Token Index: 332

Text #17

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8162. Min Act: -0.0001

Data Index: 529845 (C4 (Web Text))

Max Activating Token Index: 15

Text #18

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.7909. Min Act: -0.0001

Data Index: 1134361 (C4 (Web Text))

Max Activating Token Index: 44

Text #19

Max Range: 1.3186. Min Range: -1.3186

Max Act: 0.8102. Min Act: -0.0001

Data Index: 264879 (C4 (Web Text))

Max Activating Token Index: 62