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

Dataset: The Pile

Neuron 2883 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: 7.1721. Min Range: -7.1721

Max Act: 7.1721. Min Act: -0.0000

Data Index: 750216 (The Pile)

Max Activating Token Index: 261

Text #1

Max Range: 7.1721. Min Range: -7.1721

Max Act: 6.8345. Min Act: -0.0000

Data Index: 267031 (The Pile)

Max Activating Token Index: 288

Text #2

Max Range: 7.1721. Min Range: -7.1721

Max Act: 5.4264. Min Act: -0.0000

Data Index: 614248 (The Pile)

Max Activating Token Index: 771

Text #3

Max Range: 7.1721. Min Range: -7.1721

Max Act: 5.0840. Min Act: -0.0000

Data Index: 1654103 (The Pile)

Max Activating Token Index: 254

Text #4

Max Range: 7.1721. Min Range: -7.1721

Max Act: 5.2004. Min Act: -0.0000

Data Index: 608761 (The Pile)

Max Activating Token Index: 844

Text #5

Max Range: 7.1721. Min Range: -7.1721

Max Act: 5.0135. Min Act: -0.0000

Data Index: 1412836 (The Pile)

Max Activating Token Index: 275

Text #6

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.8908. Min Act: -0.0000

Data Index: 1357501 (The Pile)

Max Activating Token Index: 194

Text #7

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.8623. Min Act: -0.0000

Data Index: 1225066 (The Pile)

Max Activating Token Index: 343

Text #8

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.4739. Min Act: -0.0000

Data Index: 645196 (The Pile)

Max Activating Token Index: 245

Text #9

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.7758. Min Act: -0.0000

Data Index: 826694 (The Pile)

Max Activating Token Index: 684

Text #10

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.5002. Min Act: -0.0000

Data Index: 280072 (The Pile)

Max Activating Token Index: 41

Text #11

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.2650. Min Act: -0.0000

Data Index: 1106628 (The Pile)

Max Activating Token Index: 829

Text #12

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.6146. Min Act: -0.0000

Data Index: 389271 (The Pile)

Max Activating Token Index: 116

Text #13

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.5248. Min Act: -0.0000

Data Index: 625397 (The Pile)

Max Activating Token Index: 264

Text #14

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.3342. Min Act: -0.0000

Data Index: 235412 (The Pile)

Max Activating Token Index: 516

Text #15

Max Range: 7.1721. Min Range: -7.1721

Max Act: 3.9236. Min Act: -0.0000

Data Index: 736539 (The Pile)

Max Activating Token Index: 230

Text #16

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.6230. Min Act: -0.0000

Data Index: 1581581 (The Pile)

Max Activating Token Index: 688

Text #17

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.1938. Min Act: -0.0000

Data Index: 643052 (The Pile)

Max Activating Token Index: 567

Text #18

Max Range: 7.1721. Min Range: -7.1721

Max Act: 4.1707. Min Act: -0.0000

Data Index: 1681782 (The Pile)

Max Activating Token Index: 179

Text #19

Max Range: 7.1721. Min Range: -7.1721

Max Act: 3.9842. Min Act: -0.0000

Data Index: 301866 (The Pile)

Max Activating Token Index: 592