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

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

Neuron 3658 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7862. Min Act: -0.0001

Data Index: 1553283 (Python Code)

Max Activating Token Index: 309

Text #1

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.8087. Min Act: -0.0001

Data Index: 1501190 (Python Code)

Max Activating Token Index: 601

Text #2

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.8169. Min Act: -0.0001

Data Index: 244809 (C4 (Web Text))

Max Activating Token Index: 658

Text #3

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.8134. Min Act: -0.0001

Data Index: 1375190 (Python Code)

Max Activating Token Index: 839

Text #4

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7490. Min Act: -0.0001

Data Index: 814179 (C4 (Web Text))

Max Activating Token Index: 35

Text #5

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7143. Min Act: -0.0001

Data Index: 252079 (C4 (Web Text))

Max Activating Token Index: 917

Text #6

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7330. Min Act: -0.0001

Data Index: 1461999 (Python Code)

Max Activating Token Index: 1012

Text #7

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7173. Min Act: -0.0001

Data Index: 437368 (C4 (Web Text))

Max Activating Token Index: 919

Text #8

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7296. Min Act: -0.0001

Data Index: 1223455 (C4 (Web Text))

Max Activating Token Index: 940

Text #9

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7299. Min Act: -0.0001

Data Index: 1110648 (C4 (Web Text))

Max Activating Token Index: 912

Text #10

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7193. Min Act: -0.0001

Data Index: 745536 (C4 (Web Text))

Max Activating Token Index: 941

Text #11

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.6874. Min Act: -0.0001

Data Index: 1064278 (C4 (Web Text))

Max Activating Token Index: 412

Text #12

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.6582. Min Act: -0.0001

Data Index: 1619941 (Python Code)

Max Activating Token Index: 487

Text #13

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.6529. Min Act: -0.0001

Data Index: 959947 (C4 (Web Text))

Max Activating Token Index: 469

Text #14

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.6690. Min Act: -0.0001

Data Index: 922262 (C4 (Web Text))

Max Activating Token Index: 633

Text #15

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.6892. Min Act: -0.0001

Data Index: 820649 (C4 (Web Text))

Max Activating Token Index: 897

Text #16

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.7182. Min Act: -0.0001

Data Index: 672948 (C4 (Web Text))

Max Activating Token Index: 337

Text #17

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.6321. Min Act: -0.0001

Data Index: 266299 (C4 (Web Text))

Max Activating Token Index: 783

Text #18

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.6448. Min Act: -0.0001

Data Index: 713100 (C4 (Web Text))

Max Activating Token Index: 1016

Text #19

Max Range: 0.8169. Min Range: -0.8169

Max Act: 0.6657. Min Act: -0.0001

Data Index: 234045 (C4 (Web Text))

Max Activating Token Index: 839