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

Dataset: The Pile

Neuron 226 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.6884. Min Range: -5.6884

Max Act: 5.6884. Min Act: -0.0000

Data Index: 1069531 (The Pile)

Max Activating Token Index: 26

Text #1

Max Range: 5.6884. Min Range: -5.6884

Max Act: 4.5424. Min Act: -0.0000

Data Index: 1555595 (The Pile)

Max Activating Token Index: 30

Text #2

Max Range: 5.6884. Min Range: -5.6884

Max Act: 4.1714. Min Act: -0.0000

Data Index: 1503108 (The Pile)

Max Activating Token Index: 339

Text #3

Max Range: 5.6884. Min Range: -5.6884

Max Act: 4.2846. Min Act: -0.0000

Data Index: 257452 (The Pile)

Max Activating Token Index: 425

Text #4

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.9461. Min Act: -0.0000

Data Index: 1280622 (The Pile)

Max Activating Token Index: 247

Text #5

Max Range: 5.6884. Min Range: -5.6884

Max Act: 4.2681. Min Act: -0.0000

Data Index: 655072 (The Pile)

Max Activating Token Index: 25

Text #6

Max Range: 5.6884. Min Range: -5.6884

Max Act: 4.3159. Min Act: -0.0000

Data Index: 741911 (The Pile)

Max Activating Token Index: 400

Text #7

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.9490. Min Act: -0.0000

Data Index: 890004 (The Pile)

Max Activating Token Index: 464

Text #8

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.6432. Min Act: -0.0000

Data Index: 1945433 (The Pile)

Max Activating Token Index: 480

Text #9

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.5802. Min Act: -0.0000

Data Index: 1427707 (The Pile)

Max Activating Token Index: 25

Text #10

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.6325. Min Act: -0.0000

Data Index: 1576453 (The Pile)

Max Activating Token Index: 260

Text #11

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.4280. Min Act: -0.0000

Data Index: 611607 (The Pile)

Max Activating Token Index: 571

Text #12

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.7745. Min Act: -0.0000

Data Index: 953237 (The Pile)

Max Activating Token Index: 285

Text #13

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.7473. Min Act: -0.0000

Data Index: 1228673 (The Pile)

Max Activating Token Index: 190

Text #14

Max Range: 5.6884. Min Range: -5.6884

Max Act: 2.9195. Min Act: -0.0000

Data Index: 598378 (The Pile)

Max Activating Token Index: 250

Text #15

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.3345. Min Act: -0.0000

Data Index: 1006092 (The Pile)

Max Activating Token Index: 206

Text #16

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.4775. Min Act: -0.0000

Data Index: 1999852 (The Pile)

Max Activating Token Index: 371

Text #17

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.2913. Min Act: -0.0000

Data Index: 731556 (The Pile)

Max Activating Token Index: 34

Text #18

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.1384. Min Act: -0.0000

Data Index: 481535 (The Pile)

Max Activating Token Index: 185

Text #19

Max Range: 5.6884. Min Range: -5.6884

Max Act: 3.5594. Min Act: -0.0000

Data Index: 707598 (The Pile)

Max Activating Token Index: 288