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

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

Neuron 127 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.3618. Min Act: -0.1700

Data Index: 605523 (C4 (Web Text))

Max Activating Token Index: 240

Text #1

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.2680. Min Act: -0.1700

Data Index: 11608 (C4 (Web Text))

Max Activating Token Index: 276

Text #2

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.2055. Min Act: -0.1700

Data Index: 939564 (C4 (Web Text))

Max Activating Token Index: 260

Text #3

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.2055. Min Act: -0.1700

Data Index: 933501 (C4 (Web Text))

Max Activating Token Index: 17

Text #4

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.2055. Min Act: -0.1700

Data Index: 585648 (C4 (Web Text))

Max Activating Token Index: 255

Text #5

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.1743. Min Act: -0.1700

Data Index: 112291 (C4 (Web Text))

Max Activating Token Index: 38

Text #6

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.1430. Min Act: -0.1700

Data Index: 294460 (C4 (Web Text))

Max Activating Token Index: 86

Text #7

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.1430. Min Act: -0.1700

Data Index: 759124 (C4 (Web Text))

Max Activating Token Index: 672

Text #8

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.1118. Min Act: -0.1700

Data Index: 1312070 (C4 (Web Text))

Max Activating Token Index: 667

Text #9

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.1118. Min Act: -0.1700

Data Index: 570398 (C4 (Web Text))

Max Activating Token Index: 979

Text #10

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.1118. Min Act: -0.1700

Data Index: 939324 (C4 (Web Text))

Max Activating Token Index: 23

Text #11

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.1118. Min Act: -0.1700

Data Index: 208748 (C4 (Web Text))

Max Activating Token Index: 445

Text #12

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.0493. Min Act: -0.1700

Data Index: 1114147 (C4 (Web Text))

Max Activating Token Index: 361

Text #13

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.0493. Min Act: -0.1700

Data Index: 1270241 (C4 (Web Text))

Max Activating Token Index: 418

Text #14

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.0493. Min Act: -0.1700

Data Index: 877249 (C4 (Web Text))

Max Activating Token Index: 808

Text #15

Max Range: 5.3618. Min Range: -5.3618

Max Act: 4.9868. Min Act: -0.1700

Data Index: 1007111 (C4 (Web Text))

Max Activating Token Index: 36

Text #16

Max Range: 5.3618. Min Range: -5.3618

Max Act: 4.9868. Min Act: -0.1700

Data Index: 160016 (C4 (Web Text))

Max Activating Token Index: 33

Text #17

Max Range: 5.3618. Min Range: -5.3618

Max Act: 4.9868. Min Act: -0.1700

Data Index: 100438 (C4 (Web Text))

Max Activating Token Index: 267

Text #18

Max Range: 5.3618. Min Range: -5.3618

Max Act: 4.9555. Min Act: -0.1700

Data Index: 1106466 (C4 (Web Text))

Max Activating Token Index: 404

Text #19

Max Range: 5.3618. Min Range: -5.3618

Max Act: 5.0180. Min Act: -0.1700

Data Index: 804580 (C4 (Web Text))

Max Activating Token Index: 111