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

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

Neuron 2503 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3816. Min Act: -0.0001

Data Index: 848259 (C4 (Web Text))

Max Activating Token Index: 915

Text #1

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3482. Min Act: -0.0001

Data Index: 1275922 (C4 (Web Text))

Max Activating Token Index: 713

Text #2

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3224. Min Act: -0.0001

Data Index: 592057 (C4 (Web Text))

Max Activating Token Index: 936

Text #3

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3478. Min Act: -0.0001

Data Index: 607322 (C4 (Web Text))

Max Activating Token Index: 417

Text #4

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3167. Min Act: -0.0000

Data Index: 1183194 (C4 (Web Text))

Max Activating Token Index: 6

Text #5

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3284. Min Act: -0.0001

Data Index: 877224 (C4 (Web Text))

Max Activating Token Index: 930

Text #6

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.2972. Min Act: -0.0001

Data Index: 360150 (C4 (Web Text))

Max Activating Token Index: 127

Text #7

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3247. Min Act: -0.0001

Data Index: 1420054 (Python Code)

Max Activating Token Index: 333

Text #8

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3192. Min Act: -0.0001

Data Index: 1347944 (C4 (Web Text))

Max Activating Token Index: 229

Text #9

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.2957. Min Act: -0.0001

Data Index: 1212599 (C4 (Web Text))

Max Activating Token Index: 86

Text #10

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3139. Min Act: -0.0001

Data Index: 782726 (C4 (Web Text))

Max Activating Token Index: 194

Text #11

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.2919. Min Act: -0.0000

Data Index: 1108702 (C4 (Web Text))

Max Activating Token Index: 650

Text #12

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.2811. Min Act: -0.0001

Data Index: 376933 (C4 (Web Text))

Max Activating Token Index: 375

Text #13

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.2973. Min Act: -0.0001

Data Index: 676266 (C4 (Web Text))

Max Activating Token Index: 679

Text #14

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3067. Min Act: -0.0001

Data Index: 1510405 (Python Code)

Max Activating Token Index: 637

Text #15

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3082. Min Act: -0.0000

Data Index: 1294085 (C4 (Web Text))

Max Activating Token Index: 242

Text #16

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.2870. Min Act: -0.0001

Data Index: 570109 (C4 (Web Text))

Max Activating Token Index: 177

Text #17

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.2848. Min Act: -0.0001

Data Index: 1387628 (Python Code)

Max Activating Token Index: 314

Text #18

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.2935. Min Act: -0.0001

Data Index: 1133998 (C4 (Web Text))

Max Activating Token Index: 872

Text #19

Max Range: 0.3816. Min Range: -0.3816

Max Act: 0.3119. Min Act: -0.0001

Data Index: 1384281 (Python Code)

Max Activating Token Index: 511