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

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

Neuron 1498 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.3512. Min Act: -0.1700

Data Index: 279920 (C4 (Web Text))

Max Activating Token Index: 4

Text #1

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.3041. Min Act: -0.1700

Data Index: 316240 (C4 (Web Text))

Max Activating Token Index: 862

Text #2

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2883. Min Act: -0.1700

Data Index: 8539 (C4 (Web Text))

Max Activating Token Index: 185

Text #3

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2569. Min Act: -0.1700

Data Index: 341581 (C4 (Web Text))

Max Activating Token Index: 921

Text #4

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2569. Min Act: -0.1700

Data Index: 220040 (C4 (Web Text))

Max Activating Token Index: 945

Text #5

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2569. Min Act: -0.1700

Data Index: 1044345 (C4 (Web Text))

Max Activating Token Index: 97

Text #6

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2412. Min Act: -0.1700

Data Index: 91310 (C4 (Web Text))

Max Activating Token Index: 356

Text #7

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2255. Min Act: -0.1700

Data Index: 484998 (C4 (Web Text))

Max Activating Token Index: 145

Text #8

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2098. Min Act: -0.1700

Data Index: 1318987 (C4 (Web Text))

Max Activating Token Index: 13

Text #9

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2255. Min Act: -0.1700

Data Index: 980085 (C4 (Web Text))

Max Activating Token Index: 430

Text #10

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.2255. Min Act: -0.1700

Data Index: 294991 (C4 (Web Text))

Max Activating Token Index: 1003

Text #11

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1783. Min Act: -0.1700

Data Index: 1168308 (C4 (Web Text))

Max Activating Token Index: 876

Text #12

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1940. Min Act: -0.1700

Data Index: 180319 (C4 (Web Text))

Max Activating Token Index: 678

Text #13

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1940. Min Act: -0.1700

Data Index: 689677 (C4 (Web Text))

Max Activating Token Index: 243

Text #14

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1940. Min Act: -0.1700

Data Index: 1113552 (C4 (Web Text))

Max Activating Token Index: 974

Text #15

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1625. Min Act: -0.1700

Data Index: 1384696 (Python Code)

Max Activating Token Index: 5

Text #16

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1783. Min Act: -0.1700

Data Index: 1311456 (C4 (Web Text))

Max Activating Token Index: 312

Text #17

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1468. Min Act: -0.1700

Data Index: 573333 (C4 (Web Text))

Max Activating Token Index: 895

Text #18

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1625. Min Act: -0.1700

Data Index: 676883 (C4 (Web Text))

Max Activating Token Index: 847

Text #19

Max Range: 3.3512. Min Range: -3.3512

Max Act: 3.1625. Min Act: -0.1700

Data Index: 48057 (C4 (Web Text))

Max Activating Token Index: 932