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

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

Neuron 3551 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.8114. Min Act: -0.0000

Data Index: 423850 (C4 (Web Text))

Max Activating Token Index: 332

Text #1

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.6483. Min Act: -0.0000

Data Index: 667993 (C4 (Web Text))

Max Activating Token Index: 272

Text #2

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.6524. Min Act: -0.0000

Data Index: 833257 (C4 (Web Text))

Max Activating Token Index: 504

Text #3

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.6869. Min Act: -0.0000

Data Index: 951255 (C4 (Web Text))

Max Activating Token Index: 590

Text #4

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.6061. Min Act: -0.0000

Data Index: 867635 (C4 (Web Text))

Max Activating Token Index: 954

Text #5

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5552. Min Act: -0.0000

Data Index: 232849 (C4 (Web Text))

Max Activating Token Index: 775

Text #6

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5678. Min Act: -0.0000

Data Index: 728027 (C4 (Web Text))

Max Activating Token Index: 425

Text #7

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5860. Min Act: -0.0000

Data Index: 1045477 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5899. Min Act: -0.0000

Data Index: 413764 (C4 (Web Text))

Max Activating Token Index: 749

Text #9

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5818. Min Act: -0.0000

Data Index: 218570 (C4 (Web Text))

Max Activating Token Index: 120

Text #10

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5722. Min Act: -0.0000

Data Index: 509459 (C4 (Web Text))

Max Activating Token Index: 785

Text #11

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5095. Min Act: -0.0000

Data Index: 100515 (C4 (Web Text))

Max Activating Token Index: 57

Text #12

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5562. Min Act: -0.0000

Data Index: 251440 (C4 (Web Text))

Max Activating Token Index: 35

Text #13

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5393. Min Act: -0.0000

Data Index: 511946 (C4 (Web Text))

Max Activating Token Index: 724

Text #14

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5376. Min Act: -0.0000

Data Index: 1495880 (Python Code)

Max Activating Token Index: 268

Text #15

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5647. Min Act: -0.0000

Data Index: 242024 (C4 (Web Text))

Max Activating Token Index: 689

Text #16

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.4886. Min Act: -0.0000

Data Index: 20036 (C4 (Web Text))

Max Activating Token Index: 914

Text #17

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5565. Min Act: -0.0000

Data Index: 1195808 (C4 (Web Text))

Max Activating Token Index: 887

Text #18

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5113. Min Act: -0.0000

Data Index: 276760 (C4 (Web Text))

Max Activating Token Index: 200

Text #19

Max Range: 0.8114. Min Range: -0.8114

Max Act: 0.5275. Min Act: -0.0000

Data Index: 1203178 (C4 (Web Text))

Max Activating Token Index: 457