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

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

Neuron 1536 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: 5.7673. Min Range: -5.7673

Max Act: 5.7673. Min Act: -0.1700

Data Index: 1227160 (C4 (Web Text))

Max Activating Token Index: 471

Text #1

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.6110. Min Act: -0.1700

Data Index: 1185320 (C4 (Web Text))

Max Activating Token Index: 45

Text #2

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.6110. Min Act: -0.1700

Data Index: 248408 (C4 (Web Text))

Max Activating Token Index: 848

Text #3

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.7048. Min Act: -0.1700

Data Index: 644831 (C4 (Web Text))

Max Activating Token Index: 316

Text #4

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.6110. Min Act: -0.1700

Data Index: 227779 (C4 (Web Text))

Max Activating Token Index: 356

Text #5

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.6110. Min Act: -0.1700

Data Index: 315379 (C4 (Web Text))

Max Activating Token Index: 239

Text #6

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.5798. Min Act: -0.1700

Data Index: 1220898 (C4 (Web Text))

Max Activating Token Index: 633

Text #7

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.6110. Min Act: -0.1700

Data Index: 247442 (C4 (Web Text))

Max Activating Token Index: 350

Text #8

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.6110. Min Act: -0.1700

Data Index: 450751 (C4 (Web Text))

Max Activating Token Index: 252

Text #9

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4860. Min Act: -0.1700

Data Index: 520617 (C4 (Web Text))

Max Activating Token Index: 778

Text #10

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4548. Min Act: -0.1700

Data Index: 1219152 (C4 (Web Text))

Max Activating Token Index: 481

Text #11

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4860. Min Act: -0.1700

Data Index: 456219 (C4 (Web Text))

Max Activating Token Index: 230

Text #12

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4860. Min Act: -0.1700

Data Index: 1040123 (C4 (Web Text))

Max Activating Token Index: 671

Text #13

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4548. Min Act: -0.1700

Data Index: 597569 (C4 (Web Text))

Max Activating Token Index: 671

Text #14

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4235. Min Act: -0.1700

Data Index: 1309929 (C4 (Web Text))

Max Activating Token Index: 783

Text #15

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4235. Min Act: -0.1700

Data Index: 1141349 (C4 (Web Text))

Max Activating Token Index: 657

Text #16

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.3923. Min Act: -0.1700

Data Index: 344049 (C4 (Web Text))

Max Activating Token Index: 931

Text #17

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4548. Min Act: -0.1700

Data Index: 12081 (C4 (Web Text))

Max Activating Token Index: 923

Text #18

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4235. Min Act: -0.1700

Data Index: 1196357 (C4 (Web Text))

Max Activating Token Index: 984

Text #19

Max Range: 5.7673. Min Range: -5.7673

Max Act: 5.4548. Min Act: -0.1700

Data Index: 1178358 (C4 (Web Text))

Max Activating Token Index: 845