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

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

Neuron 1053 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: 0.7601. Min Range: -0.7601

Max Act: 0.7601. Min Act: -0.0001

Data Index: 291926 (C4 (Web Text))

Max Activating Token Index: 16

Text #1

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.7535. Min Act: -0.0001

Data Index: 884637 (C4 (Web Text))

Max Activating Token Index: 16

Text #2

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.7532. Min Act: -0.0001

Data Index: 531786 (C4 (Web Text))

Max Activating Token Index: 16

Text #3

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.7385. Min Act: -0.0001

Data Index: 1015975 (C4 (Web Text))

Max Activating Token Index: 24

Text #4

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.7122. Min Act: -0.0001

Data Index: 494643 (C4 (Web Text))

Max Activating Token Index: 359

Text #5

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.7290. Min Act: -0.0001

Data Index: 1582438 (Python Code)

Max Activating Token Index: 994

Text #6

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.7184. Min Act: -0.0001

Data Index: 887586 (C4 (Web Text))

Max Activating Token Index: 671

Text #7

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.7034. Min Act: -0.0001

Data Index: 980675 (C4 (Web Text))

Max Activating Token Index: 465

Text #8

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.7259. Min Act: -0.0001

Data Index: 1152606 (C4 (Web Text))

Max Activating Token Index: 13

Text #9

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6999. Min Act: -0.0001

Data Index: 265557 (C4 (Web Text))

Max Activating Token Index: 304

Text #10

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6924. Min Act: -0.0001

Data Index: 600855 (C4 (Web Text))

Max Activating Token Index: 523

Text #11

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6740. Min Act: -0.0001

Data Index: 564581 (C4 (Web Text))

Max Activating Token Index: 554

Text #12

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6839. Min Act: -0.0001

Data Index: 901211 (C4 (Web Text))

Max Activating Token Index: 68

Text #13

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6883. Min Act: -0.0001

Data Index: 1370787 (Python Code)

Max Activating Token Index: 112

Text #14

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6739. Min Act: -0.0001

Data Index: 714955 (C4 (Web Text))

Max Activating Token Index: 284

Text #15

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6706. Min Act: -0.0001

Data Index: 38141 (C4 (Web Text))

Max Activating Token Index: 515

Text #16

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6725. Min Act: -0.0001

Data Index: 1262800 (C4 (Web Text))

Max Activating Token Index: 636

Text #17

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6117. Min Act: -0.0001

Data Index: 1582144 (Python Code)

Max Activating Token Index: 35

Text #18

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6006. Min Act: -0.0001

Data Index: 1137156 (C4 (Web Text))

Max Activating Token Index: 68

Text #19

Max Range: 0.7601. Min Range: -0.7601

Max Act: 0.6177. Min Act: -0.0001

Data Index: 1131249 (C4 (Web Text))

Max Activating Token Index: 221