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

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

Neuron 2033 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.3282. Min Act: -0.0001

Data Index: 979277 (C4 (Web Text))

Max Activating Token Index: 187

Text #1

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.3066. Min Act: -0.0001

Data Index: 419528 (C4 (Web Text))

Max Activating Token Index: 196

Text #2

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.3057. Min Act: -0.0001

Data Index: 766061 (C4 (Web Text))

Max Activating Token Index: 244

Text #3

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2938. Min Act: -0.0001

Data Index: 1645469 (Python Code)

Max Activating Token Index: 134

Text #4

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.3104. Min Act: -0.0001

Data Index: 1644504 (Python Code)

Max Activating Token Index: 185

Text #5

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2994. Min Act: -0.0001

Data Index: 619826 (C4 (Web Text))

Max Activating Token Index: 178

Text #6

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2884. Min Act: -0.0001

Data Index: 1189054 (C4 (Web Text))

Max Activating Token Index: 143

Text #7

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2986. Min Act: -0.0001

Data Index: 1141073 (C4 (Web Text))

Max Activating Token Index: 213

Text #8

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2976. Min Act: -0.0001

Data Index: 52190 (C4 (Web Text))

Max Activating Token Index: 198

Text #9

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2969. Min Act: -0.0001

Data Index: 1170626 (C4 (Web Text))

Max Activating Token Index: 252

Text #10

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2868. Min Act: -0.0001

Data Index: 178477 (C4 (Web Text))

Max Activating Token Index: 156

Text #11

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2967. Min Act: -0.0001

Data Index: 48928 (C4 (Web Text))

Max Activating Token Index: 158

Text #12

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2861. Min Act: -0.0001

Data Index: 250779 (C4 (Web Text))

Max Activating Token Index: 207

Text #13

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2953. Min Act: -0.0001

Data Index: 1652066 (Python Code)

Max Activating Token Index: 176

Text #14

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2958. Min Act: -0.0001

Data Index: 59444 (C4 (Web Text))

Max Activating Token Index: 210

Text #15

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2954. Min Act: -0.0001

Data Index: 844059 (C4 (Web Text))

Max Activating Token Index: 245

Text #16

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2950. Min Act: -0.0001

Data Index: 1079651 (C4 (Web Text))

Max Activating Token Index: 227

Text #17

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2942. Min Act: -0.0001

Data Index: 407082 (C4 (Web Text))

Max Activating Token Index: 157

Text #18

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2939. Min Act: -0.0001

Data Index: 1472496 (Python Code)

Max Activating Token Index: 144

Text #19

Max Range: 0.3282. Min Range: -0.3282

Max Act: 0.2906. Min Act: -0.0001

Data Index: 534909 (C4 (Web Text))

Max Activating Token Index: 198