Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 685 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.2598. Min Act: -0.1700

Data Index: 1339726 (C4 (Web Text))

Max Activating Token Index: 898

Text #1

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.2598. Min Act: -0.1700

Data Index: 518086 (C4 (Web Text))

Max Activating Token Index: 955

Text #2

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.2598. Min Act: -0.1700

Data Index: 73773 (C4 (Web Text))

Max Activating Token Index: 27

Text #3

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.2441. Min Act: -0.1700

Data Index: 598483 (C4 (Web Text))

Max Activating Token Index: 299

Text #4

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.2441. Min Act: -0.1700

Data Index: 1142099 (C4 (Web Text))

Max Activating Token Index: 674

Text #5

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1969. Min Act: -0.1700

Data Index: 874307 (C4 (Web Text))

Max Activating Token Index: 942

Text #6

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1811. Min Act: -0.1700

Data Index: 1296690 (C4 (Web Text))

Max Activating Token Index: 572

Text #7

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.2283. Min Act: -0.1700

Data Index: 1305531 (C4 (Web Text))

Max Activating Token Index: 675

Text #8

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1811. Min Act: -0.1700

Data Index: 598292 (C4 (Web Text))

Max Activating Token Index: 355

Text #9

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1969. Min Act: -0.1700

Data Index: 331660 (C4 (Web Text))

Max Activating Token Index: 242

Text #10

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1969. Min Act: -0.1700

Data Index: 34481 (C4 (Web Text))

Max Activating Token Index: 27

Text #11

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1654. Min Act: -0.1700

Data Index: 1381820 (Python Code)

Max Activating Token Index: 545

Text #12

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1654. Min Act: -0.1700

Data Index: 1234331 (C4 (Web Text))

Max Activating Token Index: 417

Text #13

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1654. Min Act: -0.1700

Data Index: 656206 (C4 (Web Text))

Max Activating Token Index: 799

Text #14

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1339. Min Act: -0.1700

Data Index: 1153974 (C4 (Web Text))

Max Activating Token Index: 728

Text #15

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1654. Min Act: -0.1700

Data Index: 823683 (C4 (Web Text))

Max Activating Token Index: 130

Text #16

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1654. Min Act: -0.1700

Data Index: 1239093 (C4 (Web Text))

Max Activating Token Index: 792

Text #17

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1497. Min Act: -0.1700

Data Index: 1076259 (C4 (Web Text))

Max Activating Token Index: 827

Text #18

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1181. Min Act: -0.1700

Data Index: 274198 (C4 (Web Text))

Max Activating Token Index: 996

Text #19

Max Range: 3.2598. Min Range: -3.2598

Max Act: 3.1181. Min Act: -0.1700

Data Index: 468425 (C4 (Web Text))

Max Activating Token Index: 957