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

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

Neuron 1341 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: 5.6856. Min Range: -5.6856

Max Act: 5.6856. Min Act: -0.1700

Data Index: 526661 (C4 (Web Text))

Max Activating Token Index: 291

Text #1

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.6856. Min Act: -0.1700

Data Index: 1141123 (C4 (Web Text))

Max Activating Token Index: 186

Text #2

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.5919. Min Act: -0.1700

Data Index: 267582 (C4 (Web Text))

Max Activating Token Index: 693

Text #3

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.5606. Min Act: -0.1700

Data Index: 320653 (C4 (Web Text))

Max Activating Token Index: 533

Text #4

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.5294. Min Act: -0.1700

Data Index: 1408671 (Python Code)

Max Activating Token Index: 821

Text #5

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4981. Min Act: -0.1700

Data Index: 261512 (C4 (Web Text))

Max Activating Token Index: 176

Text #6

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4981. Min Act: -0.1700

Data Index: 210059 (C4 (Web Text))

Max Activating Token Index: 159

Text #7

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4981. Min Act: -0.1700

Data Index: 1341 (C4 (Web Text))

Max Activating Token Index: 158

Text #8

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4669. Min Act: -0.1700

Data Index: 295733 (C4 (Web Text))

Max Activating Token Index: 114

Text #9

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4981. Min Act: -0.1700

Data Index: 293111 (C4 (Web Text))

Max Activating Token Index: 222

Text #10

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4669. Min Act: -0.1700

Data Index: 1000034 (C4 (Web Text))

Max Activating Token Index: 287

Text #11

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4981. Min Act: -0.1700

Data Index: 776651 (C4 (Web Text))

Max Activating Token Index: 834

Text #12

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4981. Min Act: -0.1700

Data Index: 658931 (C4 (Web Text))

Max Activating Token Index: 193

Text #13

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4981. Min Act: -0.1700

Data Index: 1112153 (C4 (Web Text))

Max Activating Token Index: 231

Text #14

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4356. Min Act: -0.1700

Data Index: 99828 (C4 (Web Text))

Max Activating Token Index: 712

Text #15

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4669. Min Act: -0.1700

Data Index: 627606 (C4 (Web Text))

Max Activating Token Index: 592

Text #16

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4356. Min Act: -0.1700

Data Index: 1074753 (C4 (Web Text))

Max Activating Token Index: 685

Text #17

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.4044. Min Act: -0.1700

Data Index: 384764 (C4 (Web Text))

Max Activating Token Index: 175

Text #18

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.3731. Min Act: -0.1700

Data Index: 549795 (C4 (Web Text))

Max Activating Token Index: 556

Text #19

Max Range: 5.6856. Min Range: -5.6856

Max Act: 5.3419. Min Act: -0.1700

Data Index: 420261 (C4 (Web Text))

Max Activating Token Index: 394