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

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

Neuron 1636 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.6169. Min Act: -0.1700

Data Index: 271140 (C4 (Web Text))

Max Activating Token Index: 691

Text #1

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.5231. Min Act: -0.1700

Data Index: 1248614 (C4 (Web Text))

Max Activating Token Index: 791

Text #2

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.6169. Min Act: -0.1700

Data Index: 542718 (C4 (Web Text))

Max Activating Token Index: 337

Text #3

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.5856. Min Act: -0.1700

Data Index: 643656 (C4 (Web Text))

Max Activating Token Index: 288

Text #4

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.5856. Min Act: -0.1700

Data Index: 457215 (C4 (Web Text))

Max Activating Token Index: 496

Text #5

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4919. Min Act: -0.1700

Data Index: 161607 (C4 (Web Text))

Max Activating Token Index: 490

Text #6

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4919. Min Act: -0.1700

Data Index: 534871 (C4 (Web Text))

Max Activating Token Index: 486

Text #7

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.3981. Min Act: -0.1700

Data Index: 1003058 (C4 (Web Text))

Max Activating Token Index: 761

Text #8

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.3981. Min Act: -0.1700

Data Index: 319714 (C4 (Web Text))

Max Activating Token Index: 920

Text #9

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4294. Min Act: -0.1700

Data Index: 950172 (C4 (Web Text))

Max Activating Token Index: 242

Text #10

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4294. Min Act: -0.1700

Data Index: 877538 (C4 (Web Text))

Max Activating Token Index: 268

Text #11

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4294. Min Act: -0.1700

Data Index: 1030290 (C4 (Web Text))

Max Activating Token Index: 636

Text #12

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4294. Min Act: -0.1700

Data Index: 690859 (C4 (Web Text))

Max Activating Token Index: 532

Text #13

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.3981. Min Act: -0.1700

Data Index: 28900 (C4 (Web Text))

Max Activating Token Index: 274

Text #14

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4294. Min Act: -0.1700

Data Index: 254694 (C4 (Web Text))

Max Activating Token Index: 533

Text #15

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4919. Min Act: -0.1700

Data Index: 793616 (C4 (Web Text))

Max Activating Token Index: 372

Text #16

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.3981. Min Act: -0.1700

Data Index: 150420 (C4 (Web Text))

Max Activating Token Index: 1007

Text #17

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.4606. Min Act: -0.1700

Data Index: 788500 (C4 (Web Text))

Max Activating Token Index: 392

Text #18

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.3356. Min Act: -0.1700

Data Index: 643409 (C4 (Web Text))

Max Activating Token Index: 804

Text #19

Max Range: 5.6169. Min Range: -5.6169

Max Act: 5.3669. Min Act: -0.1700

Data Index: 314262 (C4 (Web Text))

Max Activating Token Index: 680