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

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

Neuron 5760 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.7547. Min Act: -0.0002

Data Index: 1450725 (Python Code)

Max Activating Token Index: 54

Text #1

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6804. Min Act: -0.0002

Data Index: 796828 (C4 (Web Text))

Max Activating Token Index: 11

Text #2

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6598. Min Act: -0.0002

Data Index: 713634 (C4 (Web Text))

Max Activating Token Index: 21

Text #3

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6475. Min Act: -0.0002

Data Index: 1508237 (Python Code)

Max Activating Token Index: 975

Text #4

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6212. Min Act: -0.0002

Data Index: 1378524 (Python Code)

Max Activating Token Index: 25

Text #5

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6150. Min Act: -0.0002

Data Index: 1543470 (Python Code)

Max Activating Token Index: 168

Text #6

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6128. Min Act: -0.0002

Data Index: 1092456 (C4 (Web Text))

Max Activating Token Index: 343

Text #7

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6162. Min Act: -0.0002

Data Index: 857673 (C4 (Web Text))

Max Activating Token Index: 202

Text #8

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6182. Min Act: -0.0002

Data Index: 143577 (C4 (Web Text))

Max Activating Token Index: 336

Text #9

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.6168. Min Act: -0.0002

Data Index: 1646695 (Python Code)

Max Activating Token Index: 33

Text #10

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5958. Min Act: -0.0002

Data Index: 1107944 (C4 (Web Text))

Max Activating Token Index: 4

Text #11

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5819. Min Act: -0.0002

Data Index: 1645000 (Python Code)

Max Activating Token Index: 102

Text #12

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5764. Min Act: -0.0002

Data Index: 1619328 (Python Code)

Max Activating Token Index: 91

Text #13

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5719. Min Act: -0.0002

Data Index: 1458164 (Python Code)

Max Activating Token Index: 28

Text #14

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5797. Min Act: -0.0002

Data Index: 1207054 (C4 (Web Text))

Max Activating Token Index: 192

Text #15

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5860. Min Act: -0.0002

Data Index: 477982 (C4 (Web Text))

Max Activating Token Index: 12

Text #16

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5569. Min Act: -0.0002

Data Index: 1534897 (Python Code)

Max Activating Token Index: 541

Text #17

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5797. Min Act: -0.0002

Data Index: 168026 (C4 (Web Text))

Max Activating Token Index: 16

Text #18

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5671. Min Act: -0.0002

Data Index: 1582976 (Python Code)

Max Activating Token Index: 12

Text #19

Max Range: 0.7547. Min Range: -0.7547

Max Act: 0.5741. Min Act: -0.0002

Data Index: 1297223 (C4 (Web Text))

Max Activating Token Index: 157