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

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

Neuron 24 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: 5.8913. Min Range: -5.8913

Max Act: 5.8913. Min Act: -0.1700

Data Index: 903321 (C4 (Web Text))

Max Activating Token Index: 133

Text #1

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.6725. Min Act: -0.1700

Data Index: 787591 (C4 (Web Text))

Max Activating Token Index: 126

Text #2

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.5788. Min Act: -0.1700

Data Index: 896671 (C4 (Web Text))

Max Activating Token Index: 173

Text #3

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.5788. Min Act: -0.1700

Data Index: 1309928 (C4 (Web Text))

Max Activating Token Index: 303

Text #4

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.5475. Min Act: -0.1700

Data Index: 1075755 (C4 (Web Text))

Max Activating Token Index: 188

Text #5

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.5788. Min Act: -0.1700

Data Index: 64122 (C4 (Web Text))

Max Activating Token Index: 164

Text #6

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.5163. Min Act: -0.1700

Data Index: 533794 (C4 (Web Text))

Max Activating Token Index: 338

Text #7

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.4850. Min Act: -0.1700

Data Index: 95744 (C4 (Web Text))

Max Activating Token Index: 632

Text #8

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.3913. Min Act: -0.1700

Data Index: 272506 (C4 (Web Text))

Max Activating Token Index: 593

Text #9

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.3600. Min Act: -0.1700

Data Index: 231760 (C4 (Web Text))

Max Activating Token Index: 523

Text #10

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.3913. Min Act: -0.1699

Data Index: 1023801 (C4 (Web Text))

Max Activating Token Index: 523

Text #11

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.4225. Min Act: -0.1700

Data Index: 1058497 (C4 (Web Text))

Max Activating Token Index: 672

Text #12

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.3913. Min Act: -0.1700

Data Index: 1103973 (C4 (Web Text))

Max Activating Token Index: 395

Text #13

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.3913. Min Act: -0.1700

Data Index: 242167 (C4 (Web Text))

Max Activating Token Index: 365

Text #14

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.4538. Min Act: -0.1700

Data Index: 558354 (C4 (Web Text))

Max Activating Token Index: 452

Text #15

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.3600. Min Act: -0.1700

Data Index: 1291240 (C4 (Web Text))

Max Activating Token Index: 383

Text #16

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.4538. Min Act: -0.1699

Data Index: 657120 (C4 (Web Text))

Max Activating Token Index: 842

Text #17

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.3600. Min Act: -0.1699

Data Index: 1248321 (C4 (Web Text))

Max Activating Token Index: 873

Text #18

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.2975. Min Act: -0.1700

Data Index: 110128 (C4 (Web Text))

Max Activating Token Index: 598

Text #19

Max Range: 5.8913. Min Range: -5.8913

Max Act: 5.3132. Min Act: -0.1700

Data Index: 268808 (C4 (Web Text))

Max Activating Token Index: 864