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

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

Neuron 3546 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.3159. Min Act: -0.0001

Data Index: 117292 (C4 (Web Text))

Max Activating Token Index: 511

Text #1

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.3521. Min Act: -0.0001

Data Index: 1204822 (C4 (Web Text))

Max Activating Token Index: 602

Text #2

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0463. Min Act: -0.0001

Data Index: 683197 (C4 (Web Text))

Max Activating Token Index: 773

Text #3

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0522. Min Act: -0.0001

Data Index: 1115793 (C4 (Web Text))

Max Activating Token Index: 529

Text #4

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0134. Min Act: -0.0001

Data Index: 937500 (C4 (Web Text))

Max Activating Token Index: 748

Text #5

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0122. Min Act: -0.0001

Data Index: 280502 (C4 (Web Text))

Max Activating Token Index: 874

Text #6

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0011. Min Act: -0.0001

Data Index: 1156989 (C4 (Web Text))

Max Activating Token Index: 64

Text #7

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0062. Min Act: -0.0001

Data Index: 700776 (C4 (Web Text))

Max Activating Token Index: 224

Text #8

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.7805. Min Act: -0.0001

Data Index: 634927 (C4 (Web Text))

Max Activating Token Index: 944

Text #9

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0175. Min Act: -0.0001

Data Index: 250584 (C4 (Web Text))

Max Activating Token Index: 254

Text #10

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0160. Min Act: -0.0001

Data Index: 748319 (C4 (Web Text))

Max Activating Token Index: 919

Text #11

Max Range: 3.3521. Min Range: -3.3521

Max Act: 3.0339. Min Act: -0.0001

Data Index: 1309722 (C4 (Web Text))

Max Activating Token Index: 304

Text #12

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.8454. Min Act: -0.0001

Data Index: 721429 (C4 (Web Text))

Max Activating Token Index: 251

Text #13

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.8609. Min Act: -0.0001

Data Index: 1031879 (C4 (Web Text))

Max Activating Token Index: 358

Text #14

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.6085. Min Act: -0.0001

Data Index: 696987 (C4 (Web Text))

Max Activating Token Index: 666

Text #15

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.3362. Min Act: -0.0001

Data Index: 429015 (C4 (Web Text))

Max Activating Token Index: 178

Text #16

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.3298. Min Act: -0.0001

Data Index: 937620 (C4 (Web Text))

Max Activating Token Index: 361

Text #17

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.4184. Min Act: -0.0001

Data Index: 886288 (C4 (Web Text))

Max Activating Token Index: 752

Text #18

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.2945. Min Act: -0.0001

Data Index: 661508 (C4 (Web Text))

Max Activating Token Index: 176

Text #19

Max Range: 3.3521. Min Range: -3.3521

Max Act: 2.1166. Min Act: -0.0001

Data Index: 972216 (C4 (Web Text))

Max Activating Token Index: 134