Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5412 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.4602. Min Act: -0.0000

Data Index: 756017 (C4 (Web Text))

Max Activating Token Index: 1000

Text #1

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.4300. Min Act: -0.0000

Data Index: 765468 (C4 (Web Text))

Max Activating Token Index: 142

Text #2

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.3885. Min Act: -0.0000

Data Index: 801375 (C4 (Web Text))

Max Activating Token Index: 1011

Text #3

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.3541. Min Act: -0.0000

Data Index: 383416 (C4 (Web Text))

Max Activating Token Index: 865

Text #4

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.3884. Min Act: -0.0000

Data Index: 400652 (C4 (Web Text))

Max Activating Token Index: 341

Text #5

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.3454. Min Act: -0.0000

Data Index: 405459 (C4 (Web Text))

Max Activating Token Index: 414

Text #6

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2882. Min Act: -0.0000

Data Index: 674739 (C4 (Web Text))

Max Activating Token Index: 668

Text #7

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2968. Min Act: -0.0000

Data Index: 980123 (C4 (Web Text))

Max Activating Token Index: 470

Text #8

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2694. Min Act: -0.0000

Data Index: 493115 (C4 (Web Text))

Max Activating Token Index: 982

Text #9

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2530. Min Act: -0.0000

Data Index: 476968 (C4 (Web Text))

Max Activating Token Index: 927

Text #10

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2404. Min Act: -0.0000

Data Index: 177290 (C4 (Web Text))

Max Activating Token Index: 676

Text #11

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2186. Min Act: -0.0000

Data Index: 1518604 (Python Code)

Max Activating Token Index: 329

Text #12

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2256. Min Act: -0.0000

Data Index: 439319 (C4 (Web Text))

Max Activating Token Index: 894

Text #13

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2333. Min Act: -0.0000

Data Index: 1228971 (C4 (Web Text))

Max Activating Token Index: 521

Text #14

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2363. Min Act: -0.0000

Data Index: 431581 (C4 (Web Text))

Max Activating Token Index: 321

Text #15

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2152. Min Act: -0.0000

Data Index: 138930 (C4 (Web Text))

Max Activating Token Index: 811

Text #16

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2142. Min Act: -0.0000

Data Index: 425777 (C4 (Web Text))

Max Activating Token Index: 176

Text #17

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2073. Min Act: -0.0000

Data Index: 533833 (C4 (Web Text))

Max Activating Token Index: 304

Text #18

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2204. Min Act: -0.0000

Data Index: 750566 (C4 (Web Text))

Max Activating Token Index: 947

Text #19

Max Range: 0.4602. Min Range: -0.4602

Max Act: 0.2141. Min Act: -0.0000

Data Index: 298779 (C4 (Web Text))

Max Activating Token Index: 303