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

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

Neuron 42 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: 5.5367. Min Range: -5.5367

Max Act: 5.5367. Min Act: -0.1700

Data Index: 783268 (C4 (Web Text))

Max Activating Token Index: 185

Text #1

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.4742. Min Act: -0.1700

Data Index: 132650 (C4 (Web Text))

Max Activating Token Index: 84

Text #2

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.3180. Min Act: -0.1700

Data Index: 1308224 (C4 (Web Text))

Max Activating Token Index: 213

Text #3

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.2867. Min Act: -0.1700

Data Index: 1038284 (C4 (Web Text))

Max Activating Token Index: 331

Text #4

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.2555. Min Act: -0.1700

Data Index: 1104461 (C4 (Web Text))

Max Activating Token Index: 453

Text #5

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.2242. Min Act: -0.1700

Data Index: 819197 (C4 (Web Text))

Max Activating Token Index: 467

Text #6

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.2242. Min Act: -0.1699

Data Index: 214569 (C4 (Web Text))

Max Activating Token Index: 93

Text #7

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.2242. Min Act: -0.1700

Data Index: 1129068 (C4 (Web Text))

Max Activating Token Index: 425

Text #8

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.1930. Min Act: -0.1700

Data Index: 22002 (C4 (Web Text))

Max Activating Token Index: 831

Text #9

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.1617. Min Act: -0.1700

Data Index: 743232 (C4 (Web Text))

Max Activating Token Index: 976

Text #10

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.1930. Min Act: -0.1700

Data Index: 583477 (C4 (Web Text))

Max Activating Token Index: 461

Text #11

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.2242. Min Act: -0.1700

Data Index: 1268547 (C4 (Web Text))

Max Activating Token Index: 306

Text #12

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.1617. Min Act: -0.1700

Data Index: 972156 (C4 (Web Text))

Max Activating Token Index: 90

Text #13

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.1305. Min Act: -0.1700

Data Index: 1297618 (C4 (Web Text))

Max Activating Token Index: 178

Text #14

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.1305. Min Act: -0.1700

Data Index: 687310 (C4 (Web Text))

Max Activating Token Index: 909

Text #15

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.0367. Min Act: -0.1700

Data Index: 229161 (C4 (Web Text))

Max Activating Token Index: 779

Text #16

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.0992. Min Act: -0.1700

Data Index: 661289 (C4 (Web Text))

Max Activating Token Index: 375

Text #17

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.0367. Min Act: -0.1700

Data Index: 477467 (C4 (Web Text))

Max Activating Token Index: 986

Text #18

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.0680. Min Act: -0.1700

Data Index: 1048934 (C4 (Web Text))

Max Activating Token Index: 340

Text #19

Max Range: 5.5367. Min Range: -5.5367

Max Act: 5.0680. Min Act: -0.1700

Data Index: 198005 (C4 (Web Text))

Max Activating Token Index: 262