Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4707 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.3258. Min Act: -0.0001

Data Index: 225115 (C4 (Web Text))

Max Activating Token Index: 659

Text #1

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2628. Min Act: -0.0001

Data Index: 1591980 (Python Code)

Max Activating Token Index: 328

Text #2

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2289. Min Act: -0.0001

Data Index: 257756 (C4 (Web Text))

Max Activating Token Index: 96

Text #3

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2371. Min Act: -0.0001

Data Index: 1493781 (Python Code)

Max Activating Token Index: 269

Text #4

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2131. Min Act: -0.0001

Data Index: 1403776 (Python Code)

Max Activating Token Index: 282

Text #5

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2145. Min Act: -0.0001

Data Index: 272182 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2147. Min Act: -0.0001

Data Index: 229005 (C4 (Web Text))

Max Activating Token Index: 784

Text #7

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2195. Min Act: -0.0001

Data Index: 1367504 (Python Code)

Max Activating Token Index: 116

Text #8

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2107. Min Act: -0.0001

Data Index: 981312 (C4 (Web Text))

Max Activating Token Index: 124

Text #9

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2166. Min Act: -0.0001

Data Index: 53561 (C4 (Web Text))

Max Activating Token Index: 188

Text #10

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2136. Min Act: -0.0001

Data Index: 256493 (C4 (Web Text))

Max Activating Token Index: 341

Text #11

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.1950. Min Act: -0.0001

Data Index: 199719 (C4 (Web Text))

Max Activating Token Index: 34

Text #12

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2023. Min Act: -0.0001

Data Index: 1571223 (Python Code)

Max Activating Token Index: 548

Text #13

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2180. Min Act: -0.0001

Data Index: 1491283 (Python Code)

Max Activating Token Index: 666

Text #14

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2115. Min Act: -0.0001

Data Index: 1650386 (Python Code)

Max Activating Token Index: 725

Text #15

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2006. Min Act: -0.0001

Data Index: 934550 (C4 (Web Text))

Max Activating Token Index: 290

Text #16

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.1927. Min Act: -0.0001

Data Index: 858738 (C4 (Web Text))

Max Activating Token Index: 392

Text #17

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.1883. Min Act: -0.0001

Data Index: 1512701 (Python Code)

Max Activating Token Index: 131

Text #18

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.2003. Min Act: -0.0001

Data Index: 653551 (C4 (Web Text))

Max Activating Token Index: 939

Text #19

Max Range: 0.3258. Min Range: -0.3258

Max Act: 0.1946. Min Act: -0.0001

Data Index: 1022088 (C4 (Web Text))

Max Activating Token Index: 718