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

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

Neuron 664 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4064. Min Range: -5.4064

Max Act: 5.2316. Min Act: -0.0001

Data Index: 534166 (C4 (Web Text))

Max Activating Token Index: 413

Text #1

Max Range: 5.4064. Min Range: -5.4064

Max Act: 5.4064. Min Act: -0.0001

Data Index: 758356 (C4 (Web Text))

Max Activating Token Index: 156

Text #2

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.9130. Min Act: -0.0001

Data Index: 61585 (C4 (Web Text))

Max Activating Token Index: 176

Text #3

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.8490. Min Act: -0.0001

Data Index: 788422 (C4 (Web Text))

Max Activating Token Index: 310

Text #4

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.8185. Min Act: -0.0001

Data Index: 839508 (C4 (Web Text))

Max Activating Token Index: 853

Text #5

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.9600. Min Act: -0.0001

Data Index: 427331 (C4 (Web Text))

Max Activating Token Index: 198

Text #6

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.7995. Min Act: -0.0001

Data Index: 879738 (C4 (Web Text))

Max Activating Token Index: 558

Text #7

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.6891. Min Act: -0.0001

Data Index: 98817 (C4 (Web Text))

Max Activating Token Index: 424

Text #8

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.8347. Min Act: -0.0001

Data Index: 160265 (C4 (Web Text))

Max Activating Token Index: 842

Text #9

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.5059. Min Act: -0.0001

Data Index: 949253 (C4 (Web Text))

Max Activating Token Index: 119

Text #10

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.4499. Min Act: -0.0002

Data Index: 1069160 (C4 (Web Text))

Max Activating Token Index: 720

Text #11

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.4867. Min Act: -0.0001

Data Index: 374491 (C4 (Web Text))

Max Activating Token Index: 762

Text #12

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.3059. Min Act: -0.0001

Data Index: 21010 (C4 (Web Text))

Max Activating Token Index: 788

Text #13

Max Range: 5.4064. Min Range: -5.4064

Max Act: 3.9812. Min Act: -0.0001

Data Index: 118536 (C4 (Web Text))

Max Activating Token Index: 664

Text #14

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.1311. Min Act: -0.0001

Data Index: 797186 (C4 (Web Text))

Max Activating Token Index: 891

Text #15

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.2360. Min Act: -0.0001

Data Index: 551575 (C4 (Web Text))

Max Activating Token Index: 111

Text #16

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.2328. Min Act: -0.0001

Data Index: 564055 (C4 (Web Text))

Max Activating Token Index: 209

Text #17

Max Range: 5.4064. Min Range: -5.4064

Max Act: 3.7324. Min Act: -0.0001

Data Index: 278205 (C4 (Web Text))

Max Activating Token Index: 809

Text #18

Max Range: 5.4064. Min Range: -5.4064

Max Act: 4.2034. Min Act: -0.0001

Data Index: 806880 (C4 (Web Text))

Max Activating Token Index: 965

Text #19

Max Range: 5.4064. Min Range: -5.4064

Max Act: 3.8512. Min Act: -0.0001

Data Index: 4574 (C4 (Web Text))

Max Activating Token Index: 680