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

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

Neuron 211 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: 3.3952. Min Range: -3.3952

Max Act: 3.3952. Min Act: -0.1700

Data Index: 1099089 (C4 (Web Text))

Max Activating Token Index: 123

Text #1

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2696. Min Act: -0.1700

Data Index: 1040094 (C4 (Web Text))

Max Activating Token Index: 769

Text #2

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.3167. Min Act: -0.1700

Data Index: 1000387 (C4 (Web Text))

Max Activating Token Index: 907

Text #3

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2853. Min Act: -0.1700

Data Index: 428236 (C4 (Web Text))

Max Activating Token Index: 23

Text #4

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2696. Min Act: -0.1700

Data Index: 468044 (C4 (Web Text))

Max Activating Token Index: 600

Text #5

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2539. Min Act: -0.1700

Data Index: 689181 (C4 (Web Text))

Max Activating Token Index: 716

Text #6

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2225. Min Act: -0.1700

Data Index: 931267 (C4 (Web Text))

Max Activating Token Index: 659

Text #7

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2853. Min Act: -0.1700

Data Index: 197007 (C4 (Web Text))

Max Activating Token Index: 218

Text #8

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2382. Min Act: -0.1700

Data Index: 1041035 (C4 (Web Text))

Max Activating Token Index: 127

Text #9

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2067. Min Act: -0.1700

Data Index: 983242 (C4 (Web Text))

Max Activating Token Index: 725

Text #10

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2225. Min Act: -0.1700

Data Index: 441650 (C4 (Web Text))

Max Activating Token Index: 532

Text #11

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.2225. Min Act: -0.1700

Data Index: 24336 (C4 (Web Text))

Max Activating Token Index: 485

Text #12

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.1910. Min Act: -0.1700

Data Index: 1193725 (C4 (Web Text))

Max Activating Token Index: 774

Text #13

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.1280. Min Act: -0.1700

Data Index: 639239 (C4 (Web Text))

Max Activating Token Index: 602

Text #14

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.1753. Min Act: -0.1700

Data Index: 877532 (C4 (Web Text))

Max Activating Token Index: 256

Text #15

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.1910. Min Act: -0.1700

Data Index: 845577 (C4 (Web Text))

Max Activating Token Index: 210

Text #16

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.1595. Min Act: -0.1700

Data Index: 217058 (C4 (Web Text))

Max Activating Token Index: 418

Text #17

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.1910. Min Act: -0.1700

Data Index: 845470 (C4 (Web Text))

Max Activating Token Index: 921

Text #18

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.1753. Min Act: -0.1700

Data Index: 731225 (C4 (Web Text))

Max Activating Token Index: 562

Text #19

Max Range: 3.3952. Min Range: -3.3952

Max Act: 3.1753. Min Act: -0.1700

Data Index: 45231 (C4 (Web Text))

Max Activating Token Index: 828