Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2481 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.1375. Min Act: -0.0001

Data Index: 537262 (C4 (Web Text))

Max Activating Token Index: 116

Text #1

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.1030. Min Act: -0.0001

Data Index: 1589350 (Python Code)

Max Activating Token Index: 1017

Text #2

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.1020. Min Act: -0.0001

Data Index: 1647758 (Python Code)

Max Activating Token Index: 908

Text #3

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0999. Min Act: -0.0001

Data Index: 1174171 (C4 (Web Text))

Max Activating Token Index: 741

Text #4

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.1036. Min Act: -0.0001

Data Index: 1527729 (Python Code)

Max Activating Token Index: 765

Text #5

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0987. Min Act: -0.0001

Data Index: 1064327 (C4 (Web Text))

Max Activating Token Index: 901

Text #6

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0936. Min Act: -0.0001

Data Index: 165138 (C4 (Web Text))

Max Activating Token Index: 906

Text #7

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0929. Min Act: -0.0001

Data Index: 1621036 (Python Code)

Max Activating Token Index: 851

Text #8

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0979. Min Act: -0.0001

Data Index: 654058 (C4 (Web Text))

Max Activating Token Index: 701

Text #9

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0970. Min Act: -0.0001

Data Index: 1547288 (Python Code)

Max Activating Token Index: 172

Text #10

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0868. Min Act: -0.0001

Data Index: 1006172 (C4 (Web Text))

Max Activating Token Index: 309

Text #11

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0862. Min Act: -0.0001

Data Index: 1448652 (Python Code)

Max Activating Token Index: 469

Text #12

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0893. Min Act: -0.0001

Data Index: 1549640 (Python Code)

Max Activating Token Index: 404

Text #13

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0864. Min Act: -0.0001

Data Index: 1267946 (C4 (Web Text))

Max Activating Token Index: 88

Text #14

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0916. Min Act: -0.0001

Data Index: 186910 (C4 (Web Text))

Max Activating Token Index: 139

Text #15

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0851. Min Act: -0.0001

Data Index: 937038 (C4 (Web Text))

Max Activating Token Index: 50

Text #16

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0836. Min Act: -0.0001

Data Index: 864871 (C4 (Web Text))

Max Activating Token Index: 935

Text #17

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0871. Min Act: -0.0001

Data Index: 690840 (C4 (Web Text))

Max Activating Token Index: 894

Text #18

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0863. Min Act: -0.0001

Data Index: 1297959 (C4 (Web Text))

Max Activating Token Index: 369

Text #19

Max Range: 0.1375. Min Range: -0.1375

Max Act: 0.0854. Min Act: -0.0001

Data Index: 252520 (C4 (Web Text))

Max Activating Token Index: 808