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

Dataset: The Pile

Neuron 2341 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.7303. Min Range: -4.7303

Max Act: 4.7303. Min Act: -0.0000

Data Index: 806571 (The Pile)

Max Activating Token Index: 972

Text #1

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.6980. Min Act: -0.0000

Data Index: 1308712 (The Pile)

Max Activating Token Index: 422

Text #2

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.5075. Min Act: -0.0000

Data Index: 1295908 (The Pile)

Max Activating Token Index: 101

Text #3

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.2812. Min Act: -0.0000

Data Index: 366635 (The Pile)

Max Activating Token Index: 624

Text #4

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.3302. Min Act: -0.0000

Data Index: 604925 (The Pile)

Max Activating Token Index: 1017

Text #5

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.1808. Min Act: -0.0000

Data Index: 1156019 (The Pile)

Max Activating Token Index: 576

Text #6

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.0021. Min Act: -0.0000

Data Index: 654556 (The Pile)

Max Activating Token Index: 414

Text #7

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.2784. Min Act: -0.0000

Data Index: 416385 (The Pile)

Max Activating Token Index: 67

Text #8

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.1891. Min Act: -0.0000

Data Index: 263836 (The Pile)

Max Activating Token Index: 191

Text #9

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.7218. Min Act: -0.0000

Data Index: 899209 (The Pile)

Max Activating Token Index: 561

Text #10

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.8247. Min Act: -0.0000

Data Index: 1577061 (The Pile)

Max Activating Token Index: 697

Text #11

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.9355. Min Act: -0.0000

Data Index: 654327 (The Pile)

Max Activating Token Index: 852

Text #12

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.9621. Min Act: -0.0000

Data Index: 283384 (The Pile)

Max Activating Token Index: 465

Text #13

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.8011. Min Act: -0.0000

Data Index: 1756205 (The Pile)

Max Activating Token Index: 1021

Text #14

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.6250. Min Act: -0.0000

Data Index: 1247995 (The Pile)

Max Activating Token Index: 962

Text #15

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.7797. Min Act: -0.0000

Data Index: 833292 (The Pile)

Max Activating Token Index: 901

Text #16

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.0232. Min Act: -0.0000

Data Index: 1823774 (The Pile)

Max Activating Token Index: 780

Text #17

Max Range: 4.7303. Min Range: -4.7303

Max Act: 3.0272. Min Act: -0.0000

Data Index: 1357188 (The Pile)

Max Activating Token Index: 669

Text #18

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.8712. Min Act: -0.0000

Data Index: 914626 (The Pile)

Max Activating Token Index: 56

Text #19

Max Range: 4.7303. Min Range: -4.7303

Max Act: 2.9857. Min Act: -0.0000

Data Index: 750424 (The Pile)

Max Activating Token Index: 354