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

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

Neuron 1244 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: 8.7069. Min Range: -8.7069

Max Act: 8.7069. Min Act: -0.1700

Data Index: 37105 (C4 (Web Text))

Max Activating Token Index: 390

Text #1

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.6444. Min Act: -0.1700

Data Index: 310923 (C4 (Web Text))

Max Activating Token Index: 963

Text #2

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.5507. Min Act: -0.1700

Data Index: 1358198 (C4 (Web Text))

Max Activating Token Index: 404

Text #3

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.5194. Min Act: -0.1700

Data Index: 421188 (C4 (Web Text))

Max Activating Token Index: 962

Text #4

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.4882. Min Act: -0.1700

Data Index: 510267 (C4 (Web Text))

Max Activating Token Index: 732

Text #5

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.4569. Min Act: -0.1700

Data Index: 1030357 (C4 (Web Text))

Max Activating Token Index: 352

Text #6

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.4257. Min Act: -0.1700

Data Index: 1227529 (C4 (Web Text))

Max Activating Token Index: 317

Text #7

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.4257. Min Act: -0.1700

Data Index: 552363 (C4 (Web Text))

Max Activating Token Index: 800

Text #8

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.4257. Min Act: -0.1700

Data Index: 617830 (C4 (Web Text))

Max Activating Token Index: 307

Text #9

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.2694. Min Act: -0.1700

Data Index: 457211 (C4 (Web Text))

Max Activating Token Index: 430

Text #10

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.4569. Min Act: -0.1700

Data Index: 866355 (C4 (Web Text))

Max Activating Token Index: 875

Text #11

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.4257. Min Act: -0.1700

Data Index: 722050 (C4 (Web Text))

Max Activating Token Index: 255

Text #12

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.3632. Min Act: -0.1700

Data Index: 429942 (C4 (Web Text))

Max Activating Token Index: 492

Text #13

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.3632. Min Act: -0.1700

Data Index: 103360 (C4 (Web Text))

Max Activating Token Index: 764

Text #14

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.2069. Min Act: -0.1700

Data Index: 1650610 (Python Code)

Max Activating Token Index: 256

Text #15

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.2694. Min Act: -0.1700

Data Index: 1257997 (C4 (Web Text))

Max Activating Token Index: 327

Text #16

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.3007. Min Act: -0.1700

Data Index: 7373 (C4 (Web Text))

Max Activating Token Index: 416

Text #17

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.2382. Min Act: -0.1700

Data Index: 202139 (C4 (Web Text))

Max Activating Token Index: 130

Text #18

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.2382. Min Act: -0.1700

Data Index: 716429 (C4 (Web Text))

Max Activating Token Index: 710

Text #19

Max Range: 8.7069. Min Range: -8.7069

Max Act: 8.2382. Min Act: -0.1700

Data Index: 1266362 (C4 (Web Text))

Max Activating Token Index: 292