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

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

Neuron 535 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.1194. Min Range: -7.1194

Max Act: 7.1194. Min Act: -0.1700

Data Index: 607611 (C4 (Web Text))

Max Activating Token Index: 490

Text #1

Max Range: 7.1194. Min Range: -7.1194

Max Act: 7.0257. Min Act: -0.1700

Data Index: 1004874 (C4 (Web Text))

Max Activating Token Index: 538

Text #2

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.9319. Min Act: -0.1700

Data Index: 300454 (C4 (Web Text))

Max Activating Token Index: 1020

Text #3

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.9632. Min Act: -0.1700

Data Index: 391630 (C4 (Web Text))

Max Activating Token Index: 684

Text #4

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.9319. Min Act: -0.1700

Data Index: 1231021 (C4 (Web Text))

Max Activating Token Index: 681

Text #5

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.7757. Min Act: -0.1700

Data Index: 1289030 (C4 (Web Text))

Max Activating Token Index: 459

Text #6

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.7444. Min Act: -0.1700

Data Index: 1040196 (C4 (Web Text))

Max Activating Token Index: 742

Text #7

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.7132. Min Act: -0.1700

Data Index: 140913 (C4 (Web Text))

Max Activating Token Index: 587

Text #8

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.7132. Min Act: -0.1700

Data Index: 911352 (C4 (Web Text))

Max Activating Token Index: 592

Text #9

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.6819. Min Act: -0.1700

Data Index: 585339 (C4 (Web Text))

Max Activating Token Index: 777

Text #10

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.7132. Min Act: -0.1700

Data Index: 1224045 (C4 (Web Text))

Max Activating Token Index: 518

Text #11

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.6507. Min Act: -0.1699

Data Index: 955778 (C4 (Web Text))

Max Activating Token Index: 381

Text #12

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.6194. Min Act: -0.1700

Data Index: 732925 (C4 (Web Text))

Max Activating Token Index: 196

Text #13

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.5882. Min Act: -0.1700

Data Index: 1107821 (C4 (Web Text))

Max Activating Token Index: 603

Text #14

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.5882. Min Act: -0.1700

Data Index: 468972 (C4 (Web Text))

Max Activating Token Index: 291

Text #15

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.5882. Min Act: -0.1700

Data Index: 250111 (C4 (Web Text))

Max Activating Token Index: 976

Text #16

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.6194. Min Act: -0.1700

Data Index: 623996 (C4 (Web Text))

Max Activating Token Index: 590

Text #17

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.5569. Min Act: -0.1700

Data Index: 1256960 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.5569. Min Act: -0.1700

Data Index: 622173 (C4 (Web Text))

Max Activating Token Index: 841

Text #19

Max Range: 7.1194. Min Range: -7.1194

Max Act: 6.5257. Min Act: -0.1700

Data Index: 736727 (C4 (Web Text))

Max Activating Token Index: 313