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

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

Neuron 698 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.9198. Min Act: -0.1700

Data Index: 1216874 (C4 (Web Text))

Max Activating Token Index: 242

Text #1

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.8261. Min Act: -0.1700

Data Index: 104034 (C4 (Web Text))

Max Activating Token Index: 254

Text #2

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.7948. Min Act: -0.1700

Data Index: 1096436 (C4 (Web Text))

Max Activating Token Index: 12

Text #3

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.7948. Min Act: -0.1700

Data Index: 44230 (C4 (Web Text))

Max Activating Token Index: 455

Text #4

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.7323. Min Act: -0.1700

Data Index: 257651 (C4 (Web Text))

Max Activating Token Index: 268

Text #5

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.7011. Min Act: -0.1700

Data Index: 39630 (C4 (Web Text))

Max Activating Token Index: 811

Text #6

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.7011. Min Act: -0.1700

Data Index: 1090008 (C4 (Web Text))

Max Activating Token Index: 936

Text #7

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.6698. Min Act: -0.1700

Data Index: 956051 (C4 (Web Text))

Max Activating Token Index: 25

Text #8

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.6073. Min Act: -0.1700

Data Index: 1258320 (C4 (Web Text))

Max Activating Token Index: 164

Text #9

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.5448. Min Act: -0.1700

Data Index: 871913 (C4 (Web Text))

Max Activating Token Index: 648

Text #10

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.5448. Min Act: -0.1700

Data Index: 565509 (C4 (Web Text))

Max Activating Token Index: 64

Text #11

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.5448. Min Act: -0.1700

Data Index: 23180 (C4 (Web Text))

Max Activating Token Index: 952

Text #12

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.5448. Min Act: -0.1700

Data Index: 100935 (C4 (Web Text))

Max Activating Token Index: 157

Text #13

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.4823. Min Act: -0.1700

Data Index: 41915 (C4 (Web Text))

Max Activating Token Index: 602

Text #14

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.5136. Min Act: -0.1700

Data Index: 777481 (C4 (Web Text))

Max Activating Token Index: 68

Text #15

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.5136. Min Act: -0.1699

Data Index: 715379 (C4 (Web Text))

Max Activating Token Index: 310

Text #16

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.4823. Min Act: -0.1700

Data Index: 322049 (C4 (Web Text))

Max Activating Token Index: 759

Text #17

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.5136. Min Act: -0.1700

Data Index: 311027 (C4 (Web Text))

Max Activating Token Index: 132

Text #18

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.4823. Min Act: -0.1700

Data Index: 460787 (C4 (Web Text))

Max Activating Token Index: 584

Text #19

Max Range: 5.9198. Min Range: -5.9198

Max Act: 5.4511. Min Act: -0.1700

Data Index: 115937 (C4 (Web Text))

Max Activating Token Index: 268