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

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

Neuron 3923 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2665. Min Act: -0.0000

Data Index: 1036357 (C4 (Web Text))

Max Activating Token Index: 281

Text #1

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2449. Min Act: -0.0000

Data Index: 579992 (C4 (Web Text))

Max Activating Token Index: 620

Text #2

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2369. Min Act: -0.0000

Data Index: 778214 (C4 (Web Text))

Max Activating Token Index: 332

Text #3

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2159. Min Act: -0.0000

Data Index: 966203 (C4 (Web Text))

Max Activating Token Index: 878

Text #4

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2219. Min Act: -0.0000

Data Index: 1341618 (C4 (Web Text))

Max Activating Token Index: 699

Text #5

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2152. Min Act: -0.0000

Data Index: 1209752 (C4 (Web Text))

Max Activating Token Index: 374

Text #6

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2245. Min Act: -0.0000

Data Index: 1644680 (Python Code)

Max Activating Token Index: 210

Text #7

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2166. Min Act: -0.0000

Data Index: 204875 (C4 (Web Text))

Max Activating Token Index: 665

Text #8

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2095. Min Act: -0.0000

Data Index: 467356 (C4 (Web Text))

Max Activating Token Index: 612

Text #9

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2185. Min Act: -0.0001

Data Index: 1435098 (Python Code)

Max Activating Token Index: 790

Text #10

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2077. Min Act: -0.0001

Data Index: 1492049 (Python Code)

Max Activating Token Index: 925

Text #11

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2114. Min Act: -0.0001

Data Index: 1120950 (C4 (Web Text))

Max Activating Token Index: 940

Text #12

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2103. Min Act: -0.0000

Data Index: 1641888 (Python Code)

Max Activating Token Index: 222

Text #13

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2039. Min Act: -0.0000

Data Index: 711006 (C4 (Web Text))

Max Activating Token Index: 746

Text #14

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2070. Min Act: -0.0001

Data Index: 1503532 (Python Code)

Max Activating Token Index: 362

Text #15

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2015. Min Act: -0.0000

Data Index: 778483 (C4 (Web Text))

Max Activating Token Index: 470

Text #16

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2027. Min Act: -0.0000

Data Index: 1007114 (C4 (Web Text))

Max Activating Token Index: 773

Text #17

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.2005. Min Act: -0.0000

Data Index: 1636451 (Python Code)

Max Activating Token Index: 748

Text #18

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.1994. Min Act: -0.0001

Data Index: 1139791 (C4 (Web Text))

Max Activating Token Index: 745

Text #19

Max Range: 0.2665. Min Range: -0.2665

Max Act: 0.1994. Min Act: -0.0001

Data Index: 1519090 (Python Code)

Max Activating Token Index: 572