Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 835 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.3544. Min Act: -0.1700

Data Index: 651596 (C4 (Web Text))

Max Activating Token Index: 626

Text #1

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.3544. Min Act: -0.1700

Data Index: 11900 (C4 (Web Text))

Max Activating Token Index: 783

Text #2

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.3073. Min Act: -0.1700

Data Index: 1627543 (Python Code)

Max Activating Token Index: 11

Text #3

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.3073. Min Act: -0.1700

Data Index: 198944 (C4 (Web Text))

Max Activating Token Index: 389

Text #4

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.3230. Min Act: -0.1700

Data Index: 1122779 (C4 (Web Text))

Max Activating Token Index: 616

Text #5

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2916. Min Act: -0.1700

Data Index: 524238 (C4 (Web Text))

Max Activating Token Index: 548

Text #6

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2759. Min Act: -0.1700

Data Index: 1486520 (Python Code)

Max Activating Token Index: 18

Text #7

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2916. Min Act: -0.1700

Data Index: 1003660 (C4 (Web Text))

Max Activating Token Index: 980

Text #8

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2445. Min Act: -0.1700

Data Index: 648504 (C4 (Web Text))

Max Activating Token Index: 758

Text #9

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2759. Min Act: -0.1700

Data Index: 317626 (C4 (Web Text))

Max Activating Token Index: 417

Text #10

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2445. Min Act: -0.1700

Data Index: 519988 (C4 (Web Text))

Max Activating Token Index: 879

Text #11

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2602. Min Act: -0.1700

Data Index: 1477516 (Python Code)

Max Activating Token Index: 84

Text #12

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2602. Min Act: -0.1700

Data Index: 1421643 (Python Code)

Max Activating Token Index: 15

Text #13

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2445. Min Act: -0.1700

Data Index: 602343 (C4 (Web Text))

Max Activating Token Index: 402

Text #14

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.1658. Min Act: -0.1700

Data Index: 1301513 (C4 (Web Text))

Max Activating Token Index: 452

Text #15

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.1973. Min Act: -0.1700

Data Index: 1537416 (Python Code)

Max Activating Token Index: 158

Text #16

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.1973. Min Act: -0.1700

Data Index: 1638322 (Python Code)

Max Activating Token Index: 94

Text #17

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2130. Min Act: -0.1700

Data Index: 1394714 (Python Code)

Max Activating Token Index: 100

Text #18

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2130. Min Act: -0.1700

Data Index: 55524 (C4 (Web Text))

Max Activating Token Index: 461

Text #19

Max Range: 3.3544. Min Range: -3.3544

Max Act: 3.2288. Min Act: -0.1700

Data Index: 1356362 (C4 (Web Text))

Max Activating Token Index: 463