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

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

Neuron 5 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.4308. Min Act: -0.1700

Data Index: 645586 (C4 (Web Text))

Max Activating Token Index: 720

Text #1

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.3995. Min Act: -0.1700

Data Index: 570700 (C4 (Web Text))

Max Activating Token Index: 576

Text #2

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.3053. Min Act: -0.1700

Data Index: 333498 (C4 (Web Text))

Max Activating Token Index: 780

Text #3

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2896. Min Act: -0.1700

Data Index: 327864 (C4 (Web Text))

Max Activating Token Index: 843

Text #4

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2739. Min Act: -0.1700

Data Index: 1178593 (C4 (Web Text))

Max Activating Token Index: 340

Text #5

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2739. Min Act: -0.1700

Data Index: 946601 (C4 (Web Text))

Max Activating Token Index: 148

Text #6

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2582. Min Act: -0.1700

Data Index: 98185 (C4 (Web Text))

Max Activating Token Index: 21

Text #7

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2739. Min Act: -0.1700

Data Index: 1028343 (C4 (Web Text))

Max Activating Token Index: 690

Text #8

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2582. Min Act: -0.1700

Data Index: 1035351 (C4 (Web Text))

Max Activating Token Index: 692

Text #9

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2425. Min Act: -0.1700

Data Index: 267749 (C4 (Web Text))

Max Activating Token Index: 107

Text #10

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2267. Min Act: -0.1700

Data Index: 1141404 (C4 (Web Text))

Max Activating Token Index: 565

Text #11

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2110. Min Act: -0.1700

Data Index: 848196 (C4 (Web Text))

Max Activating Token Index: 772

Text #12

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2267. Min Act: -0.1700

Data Index: 551952 (C4 (Web Text))

Max Activating Token Index: 208

Text #13

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2110. Min Act: -0.1700

Data Index: 762414 (C4 (Web Text))

Max Activating Token Index: 885

Text #14

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2267. Min Act: -0.1700

Data Index: 1335360 (C4 (Web Text))

Max Activating Token Index: 354

Text #15

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2267. Min Act: -0.1700

Data Index: 597787 (C4 (Web Text))

Max Activating Token Index: 215

Text #16

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2110. Min Act: -0.1700

Data Index: 89091 (C4 (Web Text))

Max Activating Token Index: 993

Text #17

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2110. Min Act: -0.1700

Data Index: 1151061 (C4 (Web Text))

Max Activating Token Index: 475

Text #18

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.2267. Min Act: -0.1700

Data Index: 427069 (C4 (Web Text))

Max Activating Token Index: 424

Text #19

Max Range: 3.4308. Min Range: -3.4308

Max Act: 3.1953. Min Act: -0.1700

Data Index: 774142 (C4 (Web Text))

Max Activating Token Index: 570