Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1848 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.1069. Min Range: -5.1069

Max Act: 5.1069. Min Act: -0.1700

Data Index: 533597 (C4 (Web Text))

Max Activating Token Index: 123

Text #1

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.9507. Min Act: -0.1700

Data Index: 40617 (C4 (Web Text))

Max Activating Token Index: 594

Text #2

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.9194. Min Act: -0.1700

Data Index: 53776 (C4 (Web Text))

Max Activating Token Index: 492

Text #3

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.8257. Min Act: -0.1687

Data Index: 886581 (C4 (Web Text))

Max Activating Token Index: 990

Text #4

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.8569. Min Act: -0.1700

Data Index: 677576 (C4 (Web Text))

Max Activating Token Index: 167

Text #5

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.8257. Min Act: -0.1700

Data Index: 1264639 (C4 (Web Text))

Max Activating Token Index: 70

Text #6

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7944. Min Act: -0.1700

Data Index: 223346 (C4 (Web Text))

Max Activating Token Index: 831

Text #7

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7944. Min Act: -0.1699

Data Index: 638336 (C4 (Web Text))

Max Activating Token Index: 14

Text #8

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7319. Min Act: -0.1700

Data Index: 1295596 (C4 (Web Text))

Max Activating Token Index: 804

Text #9

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7944. Min Act: -0.1700

Data Index: 506220 (C4 (Web Text))

Max Activating Token Index: 289

Text #10

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.8257. Min Act: -0.1700

Data Index: 156089 (C4 (Web Text))

Max Activating Token Index: 217

Text #11

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.8257. Min Act: -0.1700

Data Index: 195657 (C4 (Web Text))

Max Activating Token Index: 383

Text #12

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7632. Min Act: -0.1700

Data Index: 950791 (C4 (Web Text))

Max Activating Token Index: 739

Text #13

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7319. Min Act: -0.1700

Data Index: 866785 (C4 (Web Text))

Max Activating Token Index: 465

Text #14

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7319. Min Act: -0.1700

Data Index: 308173 (C4 (Web Text))

Max Activating Token Index: 761

Text #15

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7007. Min Act: -0.1700

Data Index: 878434 (C4 (Web Text))

Max Activating Token Index: 801

Text #16

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7319. Min Act: -0.1700

Data Index: 677620 (C4 (Web Text))

Max Activating Token Index: 449

Text #17

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7007. Min Act: -0.1700

Data Index: 1104943 (C4 (Web Text))

Max Activating Token Index: 411

Text #18

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.6694. Min Act: -0.1700

Data Index: 182943 (C4 (Web Text))

Max Activating Token Index: 993

Text #19

Max Range: 5.1069. Min Range: -5.1069

Max Act: 4.7007. Min Act: -0.1700

Data Index: 963677 (C4 (Web Text))

Max Activating Token Index: 931