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

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

Neuron 1825 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.1154. Min Range: -7.1154

Max Act: 7.1154. Min Act: -0.1700

Data Index: 713981 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.9279. Min Act: -0.1700

Data Index: 274174 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.4904. Min Act: -0.1700

Data Index: 965020 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.5529. Min Act: -0.1700

Data Index: 296789 (C4 (Web Text))

Max Activating Token Index: 509

Text #4

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.4279. Min Act: -0.1700

Data Index: 318652 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.6467. Min Act: -0.1699

Data Index: 970889 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.3967. Min Act: -0.1700

Data Index: 963629 (C4 (Web Text))

Max Activating Token Index: 807

Text #7

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.4279. Min Act: -0.1700

Data Index: 930733 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.4279. Min Act: -0.1700

Data Index: 275932 (C4 (Web Text))

Max Activating Token Index: 607

Text #9

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.3654. Min Act: -0.1700

Data Index: 612713 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.4592. Min Act: -0.1700

Data Index: 1221771 (C4 (Web Text))

Max Activating Token Index: 479

Text #11

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.3029. Min Act: -0.1700

Data Index: 128889 (C4 (Web Text))

Max Activating Token Index: 545

Text #12

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.3029. Min Act: -0.1700

Data Index: 445567 (C4 (Web Text))

Max Activating Token Index: 134

Text #13

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.4592. Min Act: -0.1700

Data Index: 897587 (C4 (Web Text))

Max Activating Token Index: 594

Text #14

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.3342. Min Act: -0.1700

Data Index: 65032 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.3342. Min Act: -0.1700

Data Index: 1591842 (Python Code)

Max Activating Token Index: 481

Text #16

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.2092. Min Act: -0.1699

Data Index: 965674 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.4279. Min Act: -0.1700

Data Index: 143934 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.3029. Min Act: -0.1700

Data Index: 902341 (C4 (Web Text))

Max Activating Token Index: 904

Text #19

Max Range: 7.1154. Min Range: -7.1154

Max Act: 6.3342. Min Act: -0.1699

Data Index: 1329994 (C4 (Web Text))

Max Activating Token Index: 577