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

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

Neuron 2685 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: 0.1367. Min Range: -0.1367

Max Act: 0.1367. Min Act: -0.0000

Data Index: 109137 (C4 (Web Text))

Max Activating Token Index: 933

Text #1

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1246. Min Act: -0.0000

Data Index: 1254947 (C4 (Web Text))

Max Activating Token Index: 164

Text #2

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1250. Min Act: -0.0000

Data Index: 1283333 (C4 (Web Text))

Max Activating Token Index: 1000

Text #3

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1232. Min Act: -0.0000

Data Index: 1236222 (C4 (Web Text))

Max Activating Token Index: 792

Text #4

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1161. Min Act: -0.0000

Data Index: 260600 (C4 (Web Text))

Max Activating Token Index: 269

Text #5

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1073. Min Act: -0.0000

Data Index: 108080 (C4 (Web Text))

Max Activating Token Index: 976

Text #6

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1099. Min Act: -0.0001

Data Index: 801667 (C4 (Web Text))

Max Activating Token Index: 740

Text #7

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1091. Min Act: -0.0000

Data Index: 289270 (C4 (Web Text))

Max Activating Token Index: 1015

Text #8

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1051. Min Act: -0.0000

Data Index: 1126946 (C4 (Web Text))

Max Activating Token Index: 591

Text #9

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0996. Min Act: -0.0000

Data Index: 390344 (C4 (Web Text))

Max Activating Token Index: 225

Text #10

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0969. Min Act: -0.0000

Data Index: 382194 (C4 (Web Text))

Max Activating Token Index: 884

Text #11

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1007. Min Act: -0.0000

Data Index: 450115 (C4 (Web Text))

Max Activating Token Index: 312

Text #12

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.1026. Min Act: -0.0000

Data Index: 207693 (C4 (Web Text))

Max Activating Token Index: 959

Text #13

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0958. Min Act: -0.0000

Data Index: 983741 (C4 (Web Text))

Max Activating Token Index: 100

Text #14

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0993. Min Act: -0.0000

Data Index: 525321 (C4 (Web Text))

Max Activating Token Index: 942

Text #15

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0985. Min Act: -0.0000

Data Index: 982920 (C4 (Web Text))

Max Activating Token Index: 574

Text #16

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0976. Min Act: -0.0001

Data Index: 1467674 (Python Code)

Max Activating Token Index: 914

Text #17

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0965. Min Act: -0.0000

Data Index: 152906 (C4 (Web Text))

Max Activating Token Index: 827

Text #18

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0996. Min Act: -0.0000

Data Index: 1079516 (C4 (Web Text))

Max Activating Token Index: 295

Text #19

Max Range: 0.1367. Min Range: -0.1367

Max Act: 0.0957. Min Act: -0.0001

Data Index: 111905 (C4 (Web Text))

Max Activating Token Index: 1004