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

Dataset: The Pile

Neuron 4433 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.7254. Min Act: -0.0000

Data Index: 158764 (The Pile)

Max Activating Token Index: 458

Text #1

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.5248. Min Act: -0.0001

Data Index: 1019238 (The Pile)

Max Activating Token Index: 267

Text #2

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.3156. Min Act: -0.0000

Data Index: 1152615 (The Pile)

Max Activating Token Index: 506

Text #3

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.2589. Min Act: -0.0000

Data Index: 8311 (The Pile)

Max Activating Token Index: 705

Text #4

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.2253. Min Act: -0.0000

Data Index: 1642833 (The Pile)

Max Activating Token Index: 844

Text #5

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.0561. Min Act: -0.0000

Data Index: 1743419 (The Pile)

Max Activating Token Index: 449

Text #6

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.1834. Min Act: -0.0000

Data Index: 1977344 (The Pile)

Max Activating Token Index: 674

Text #7

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.2437. Min Act: -0.0000

Data Index: 286991 (The Pile)

Max Activating Token Index: 639

Text #8

Max Range: 10.7254. Min Range: -10.7254

Max Act: 10.0080. Min Act: -0.0000

Data Index: 1577335 (The Pile)

Max Activating Token Index: 359

Text #9

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.5299. Min Act: -0.0000

Data Index: 257996 (The Pile)

Max Activating Token Index: 798

Text #10

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.7188. Min Act: -0.0000

Data Index: 868852 (The Pile)

Max Activating Token Index: 520

Text #11

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.7146. Min Act: -0.0000

Data Index: 361039 (The Pile)

Max Activating Token Index: 163

Text #12

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.6585. Min Act: -0.0000

Data Index: 1347327 (The Pile)

Max Activating Token Index: 846

Text #13

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.8439. Min Act: -0.0000

Data Index: 1892835 (The Pile)

Max Activating Token Index: 788

Text #14

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.7681. Min Act: -0.0000

Data Index: 1911060 (The Pile)

Max Activating Token Index: 523

Text #15

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.6620. Min Act: -0.0000

Data Index: 906214 (The Pile)

Max Activating Token Index: 171

Text #16

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.7664. Min Act: -0.0000

Data Index: 1153095 (The Pile)

Max Activating Token Index: 484

Text #17

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.2364. Min Act: -0.0000

Data Index: 1873220 (The Pile)

Max Activating Token Index: 611

Text #18

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.4336. Min Act: -0.0000

Data Index: 1287486 (The Pile)

Max Activating Token Index: 75

Text #19

Max Range: 10.7254. Min Range: -10.7254

Max Act: 9.7371. Min Act: -0.0000

Data Index: 1933076 (The Pile)

Max Activating Token Index: 277