Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3082 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2722. Min Act: -0.0000

Data Index: 112945 (C4 (Web Text))

Max Activating Token Index: 228

Text #1

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2740. Min Act: -0.0000

Data Index: 391567 (C4 (Web Text))

Max Activating Token Index: 226

Text #2

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2511. Min Act: 0.0000

Data Index: 77275 (C4 (Web Text))

Max Activating Token Index: 799

Text #3

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2441. Min Act: -0.0000

Data Index: 324400 (C4 (Web Text))

Max Activating Token Index: 192

Text #4

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2368. Min Act: -0.0000

Data Index: 8656 (C4 (Web Text))

Max Activating Token Index: 529

Text #5

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2272. Min Act: -0.0000

Data Index: 979262 (C4 (Web Text))

Max Activating Token Index: 205

Text #6

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2271. Min Act: 0.0000

Data Index: 318949 (C4 (Web Text))

Max Activating Token Index: 894

Text #7

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2186. Min Act: -0.0000

Data Index: 1108732 (C4 (Web Text))

Max Activating Token Index: 856

Text #8

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2409. Min Act: 0.0000

Data Index: 1140965 (C4 (Web Text))

Max Activating Token Index: 662

Text #9

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2297. Min Act: -0.0000

Data Index: 528605 (C4 (Web Text))

Max Activating Token Index: 841

Text #10

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2148. Min Act: -0.0000

Data Index: 386935 (C4 (Web Text))

Max Activating Token Index: 990

Text #11

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2086. Min Act: -0.0000

Data Index: 942584 (C4 (Web Text))

Max Activating Token Index: 356

Text #12

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2146. Min Act: -0.0000

Data Index: 588638 (C4 (Web Text))

Max Activating Token Index: 437

Text #13

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2208. Min Act: -0.0000

Data Index: 707004 (C4 (Web Text))

Max Activating Token Index: 492

Text #14

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2037. Min Act: -0.0000

Data Index: 1003695 (C4 (Web Text))

Max Activating Token Index: 786

Text #15

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2203. Min Act: -0.0000

Data Index: 738541 (C4 (Web Text))

Max Activating Token Index: 208

Text #16

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2177. Min Act: -0.0000

Data Index: 3720 (C4 (Web Text))

Max Activating Token Index: 772

Text #17

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2091. Min Act: -0.0000

Data Index: 81650 (C4 (Web Text))

Max Activating Token Index: 846

Text #18

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2087. Min Act: -0.0000

Data Index: 1207591 (C4 (Web Text))

Max Activating Token Index: 976

Text #19

Max Range: 0.2740. Min Range: -0.2740

Max Act: 0.2138. Min Act: -0.0000

Data Index: 1137366 (C4 (Web Text))

Max Activating Token Index: 824