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

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

Neuron 1823 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0641. Min Range: -1.0641

Max Act: 1.0328. Min Act: -0.0001

Data Index: 464748 (C4 (Web Text))

Max Activating Token Index: 83

Text #1

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.9653. Min Act: -0.0001

Data Index: 1127756 (C4 (Web Text))

Max Activating Token Index: 340

Text #2

Max Range: 1.0641. Min Range: -1.0641

Max Act: 1.0090. Min Act: -0.0001

Data Index: 1369062 (Python Code)

Max Activating Token Index: 61

Text #3

Max Range: 1.0641. Min Range: -1.0641

Max Act: 1.0029. Min Act: -0.0001

Data Index: 200665 (C4 (Web Text))

Max Activating Token Index: 240

Text #4

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.9305. Min Act: -0.0001

Data Index: 423253 (C4 (Web Text))

Max Activating Token Index: 180

Text #5

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.9078. Min Act: -0.0001

Data Index: 1436512 (Python Code)

Max Activating Token Index: 983

Text #6

Max Range: 1.0641. Min Range: -1.0641

Max Act: 1.0641. Min Act: -0.0001

Data Index: 1372670 (Python Code)

Max Activating Token Index: 210

Text #7

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.9304. Min Act: -0.0001

Data Index: 888055 (C4 (Web Text))

Max Activating Token Index: 185

Text #8

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.9171. Min Act: -0.0001

Data Index: 1351517 (C4 (Web Text))

Max Activating Token Index: 158

Text #9

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.8585. Min Act: -0.0001

Data Index: 556157 (C4 (Web Text))

Max Activating Token Index: 191

Text #10

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.8673. Min Act: -0.0001

Data Index: 431683 (C4 (Web Text))

Max Activating Token Index: 931

Text #11

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.7717. Min Act: -0.0001

Data Index: 1543973 (Python Code)

Max Activating Token Index: 57

Text #12

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.8422. Min Act: -0.0001

Data Index: 741984 (C4 (Web Text))

Max Activating Token Index: 15

Text #13

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.8127. Min Act: -0.0001

Data Index: 1415940 (Python Code)

Max Activating Token Index: 48

Text #14

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.8879. Min Act: -0.0001

Data Index: 1349000 (C4 (Web Text))

Max Activating Token Index: 84

Text #15

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.8659. Min Act: -0.0001

Data Index: 783614 (C4 (Web Text))

Max Activating Token Index: 149

Text #16

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.9106. Min Act: -0.0001

Data Index: 1563605 (Python Code)

Max Activating Token Index: 162

Text #17

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.8841. Min Act: -0.0001

Data Index: 1608059 (Python Code)

Max Activating Token Index: 36

Text #18

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.8507. Min Act: -0.0001

Data Index: 1136982 (C4 (Web Text))

Max Activating Token Index: 140

Text #19

Max Range: 1.0641. Min Range: -1.0641

Max Act: 0.7765. Min Act: -0.0001

Data Index: 1258221 (C4 (Web Text))

Max Activating Token Index: 369