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

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

Neuron 2085 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.9163. Min Act: -0.0000

Data Index: 1588531 (Python Code)

Max Activating Token Index: 14

Text #1

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.8866. Min Act: -0.0000

Data Index: 844492 (C4 (Web Text))

Max Activating Token Index: 259

Text #2

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.8678. Min Act: -0.0000

Data Index: 688736 (C4 (Web Text))

Max Activating Token Index: 495

Text #3

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.7930. Min Act: -0.0000

Data Index: 1577307 (Python Code)

Max Activating Token Index: 11

Text #4

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.7809. Min Act: -0.0000

Data Index: 1568535 (Python Code)

Max Activating Token Index: 10

Text #5

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.7536. Min Act: -0.0000

Data Index: 1638696 (Python Code)

Max Activating Token Index: 524

Text #6

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.7256. Min Act: -0.0000

Data Index: 333029 (C4 (Web Text))

Max Activating Token Index: 238

Text #7

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.5677. Min Act: -0.0000

Data Index: 1579948 (Python Code)

Max Activating Token Index: 543

Text #8

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.8036. Min Act: -0.0000

Data Index: 131728 (C4 (Web Text))

Max Activating Token Index: 902

Text #9

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.6557. Min Act: -0.0000

Data Index: 713084 (C4 (Web Text))

Max Activating Token Index: 130

Text #10

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.5034. Min Act: -0.0000

Data Index: 1551991 (Python Code)

Max Activating Token Index: 74

Text #11

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.7034. Min Act: -0.0000

Data Index: 1428458 (Python Code)

Max Activating Token Index: 307

Text #12

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.4313. Min Act: -0.0000

Data Index: 801192 (C4 (Web Text))

Max Activating Token Index: 541

Text #13

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.5634. Min Act: -0.0000

Data Index: 556578 (C4 (Web Text))

Max Activating Token Index: 1017

Text #14

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.5576. Min Act: -0.0000

Data Index: 1557844 (Python Code)

Max Activating Token Index: 178

Text #15

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.6560. Min Act: -0.0000

Data Index: 1245442 (C4 (Web Text))

Max Activating Token Index: 840

Text #16

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.5162. Min Act: -0.0000

Data Index: 1115193 (C4 (Web Text))

Max Activating Token Index: 1003

Text #17

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.5798. Min Act: -0.0000

Data Index: 1364667 (Python Code)

Max Activating Token Index: 790

Text #18

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.4761. Min Act: -0.0000

Data Index: 1131332 (C4 (Web Text))

Max Activating Token Index: 823

Text #19

Max Range: 2.9163. Min Range: -2.9163

Max Act: 2.4727. Min Act: -0.0000

Data Index: 43706 (C4 (Web Text))

Max Activating Token Index: 368