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

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

Neuron 1673 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.5954. Min Act: -0.1700

Data Index: 378821 (C4 (Web Text))

Max Activating Token Index: 975

Text #1

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.5484. Min Act: -0.1700

Data Index: 465295 (C4 (Web Text))

Max Activating Token Index: 246

Text #2

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.4544. Min Act: -0.1700

Data Index: 691993 (C4 (Web Text))

Max Activating Token Index: 411

Text #3

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.4544. Min Act: -0.1700

Data Index: 222279 (C4 (Web Text))

Max Activating Token Index: 668

Text #4

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.4387. Min Act: -0.1700

Data Index: 973023 (C4 (Web Text))

Max Activating Token Index: 845

Text #5

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3916. Min Act: -0.1700

Data Index: 954413 (C4 (Web Text))

Max Activating Token Index: 854

Text #6

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3916. Min Act: -0.1700

Data Index: 1051533 (C4 (Web Text))

Max Activating Token Index: 995

Text #7

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.4073. Min Act: -0.1700

Data Index: 1013033 (C4 (Web Text))

Max Activating Token Index: 796

Text #8

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3760. Min Act: -0.1700

Data Index: 450452 (C4 (Web Text))

Max Activating Token Index: 487

Text #9

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3446. Min Act: -0.1700

Data Index: 315246 (C4 (Web Text))

Max Activating Token Index: 653

Text #10

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3446. Min Act: -0.1700

Data Index: 1208911 (C4 (Web Text))

Max Activating Token Index: 350

Text #11

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3603. Min Act: -0.1700

Data Index: 991671 (C4 (Web Text))

Max Activating Token Index: 558

Text #12

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3289. Min Act: -0.1700

Data Index: 1088659 (C4 (Web Text))

Max Activating Token Index: 107

Text #13

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3603. Min Act: -0.1700

Data Index: 971967 (C4 (Web Text))

Max Activating Token Index: 33

Text #14

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3289. Min Act: -0.1700

Data Index: 178911 (C4 (Web Text))

Max Activating Token Index: 280

Text #15

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3289. Min Act: -0.1700

Data Index: 724334 (C4 (Web Text))

Max Activating Token Index: 288

Text #16

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3132. Min Act: -0.1700

Data Index: 26430 (C4 (Web Text))

Max Activating Token Index: 288

Text #17

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.2975. Min Act: -0.1700

Data Index: 224890 (C4 (Web Text))

Max Activating Token Index: 28

Text #18

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3132. Min Act: -0.1700

Data Index: 658361 (C4 (Web Text))

Max Activating Token Index: 477

Text #19

Max Range: 3.5954. Min Range: -3.5954

Max Act: 3.3132. Min Act: -0.1700

Data Index: 535918 (C4 (Web Text))

Max Activating Token Index: 849