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

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

Neuron 72 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: 5.4243. Min Range: -5.4243

Max Act: 5.4243. Min Act: -0.1700

Data Index: 867088 (C4 (Web Text))

Max Activating Token Index: 695

Text #1

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.3618. Min Act: -0.1700

Data Index: 590903 (C4 (Web Text))

Max Activating Token Index: 987

Text #2

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.2368. Min Act: -0.1700

Data Index: 17204 (C4 (Web Text))

Max Activating Token Index: 54

Text #3

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1743. Min Act: -0.1700

Data Index: 572258 (C4 (Web Text))

Max Activating Token Index: 415

Text #4

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1743. Min Act: -0.1700

Data Index: 393925 (C4 (Web Text))

Max Activating Token Index: 304

Text #5

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1431. Min Act: -0.1700

Data Index: 561579 (C4 (Web Text))

Max Activating Token Index: 355

Text #6

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1118. Min Act: -0.1700

Data Index: 869158 (C4 (Web Text))

Max Activating Token Index: 314

Text #7

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1743. Min Act: -0.1700

Data Index: 1268906 (C4 (Web Text))

Max Activating Token Index: 328

Text #8

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1431. Min Act: -0.1700

Data Index: 117677 (C4 (Web Text))

Max Activating Token Index: 270

Text #9

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.0806. Min Act: -0.1700

Data Index: 1185611 (C4 (Web Text))

Max Activating Token Index: 441

Text #10

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1431. Min Act: -0.1700

Data Index: 187981 (C4 (Web Text))

Max Activating Token Index: 372

Text #11

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1118. Min Act: -0.1700

Data Index: 806052 (C4 (Web Text))

Max Activating Token Index: 444

Text #12

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1743. Min Act: -0.1700

Data Index: 462532 (C4 (Web Text))

Max Activating Token Index: 53

Text #13

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.0806. Min Act: -0.1700

Data Index: 915232 (C4 (Web Text))

Max Activating Token Index: 409

Text #14

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.0806. Min Act: -0.1700

Data Index: 108359 (C4 (Web Text))

Max Activating Token Index: 676

Text #15

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.0493. Min Act: -0.1700

Data Index: 381590 (C4 (Web Text))

Max Activating Token Index: 595

Text #16

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1118. Min Act: -0.1700

Data Index: 1335301 (C4 (Web Text))

Max Activating Token Index: 49

Text #17

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.1118. Min Act: -0.1700

Data Index: 207154 (C4 (Web Text))

Max Activating Token Index: 886

Text #18

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.0181. Min Act: -0.1700

Data Index: 748916 (C4 (Web Text))

Max Activating Token Index: 413

Text #19

Max Range: 5.4243. Min Range: -5.4243

Max Act: 5.0806. Min Act: -0.1700

Data Index: 953245 (C4 (Web Text))

Max Activating Token Index: 344