Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 521 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2288. Min Act: -0.0001

Data Index: 224901 (C4 (Web Text))

Max Activating Token Index: 832

Text #1

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2340. Min Act: -0.0001

Data Index: 1492248 (Python Code)

Max Activating Token Index: 711

Text #2

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2400. Min Act: -0.0001

Data Index: 970622 (C4 (Web Text))

Max Activating Token Index: 552

Text #3

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2377. Min Act: -0.0001

Data Index: 1417833 (Python Code)

Max Activating Token Index: 416

Text #4

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2133. Min Act: -0.0001

Data Index: 694419 (C4 (Web Text))

Max Activating Token Index: 580

Text #5

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2281. Min Act: -0.0001

Data Index: 1150274 (C4 (Web Text))

Max Activating Token Index: 402

Text #6

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2123. Min Act: -0.0001

Data Index: 1036421 (C4 (Web Text))

Max Activating Token Index: 486

Text #7

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2183. Min Act: -0.0001

Data Index: 1132367 (C4 (Web Text))

Max Activating Token Index: 846

Text #8

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2295. Min Act: -0.0001

Data Index: 419381 (C4 (Web Text))

Max Activating Token Index: 247

Text #9

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2112. Min Act: -0.0000

Data Index: 1390172 (Python Code)

Max Activating Token Index: 718

Text #10

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2067. Min Act: -0.0001

Data Index: 1131113 (C4 (Web Text))

Max Activating Token Index: 644

Text #11

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2197. Min Act: -0.0001

Data Index: 994215 (C4 (Web Text))

Max Activating Token Index: 323

Text #12

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2173. Min Act: -0.0001

Data Index: 249454 (C4 (Web Text))

Max Activating Token Index: 1011

Text #13

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2080. Min Act: -0.0001

Data Index: 734145 (C4 (Web Text))

Max Activating Token Index: 340

Text #14

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2211. Min Act: -0.0001

Data Index: 1237451 (C4 (Web Text))

Max Activating Token Index: 816

Text #15

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2010. Min Act: -0.0000

Data Index: 1641520 (Python Code)

Max Activating Token Index: 458

Text #16

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2069. Min Act: -0.0001

Data Index: 115087 (C4 (Web Text))

Max Activating Token Index: 795

Text #17

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2216. Min Act: -0.0001

Data Index: 783976 (C4 (Web Text))

Max Activating Token Index: 641

Text #18

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.2061. Min Act: -0.0001

Data Index: 61183 (C4 (Web Text))

Max Activating Token Index: 44

Text #19

Max Range: 0.2400. Min Range: -0.2400

Max Act: 0.1865. Min Act: -0.0001

Data Index: 165402 (C4 (Web Text))

Max Activating Token Index: 649