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

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

Neuron 897 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: 5.0007. Min Range: -5.0007

Max Act: 4.9695. Min Act: -0.1700

Data Index: 588625 (C4 (Web Text))

Max Activating Token Index: 155

Text #1

Max Range: 5.0007. Min Range: -5.0007

Max Act: 5.0007. Min Act: -0.1700

Data Index: 1022871 (C4 (Web Text))

Max Activating Token Index: 218

Text #2

Max Range: 5.0007. Min Range: -5.0007

Max Act: 5.0007. Min Act: -0.1700

Data Index: 948338 (C4 (Web Text))

Max Activating Token Index: 269

Text #3

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.9382. Min Act: -0.1700

Data Index: 1368661 (Python Code)

Max Activating Token Index: 709

Text #4

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.9070. Min Act: -0.1700

Data Index: 633749 (C4 (Web Text))

Max Activating Token Index: 699

Text #5

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.9070. Min Act: -0.1700

Data Index: 936305 (C4 (Web Text))

Max Activating Token Index: 157

Text #6

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.9382. Min Act: -0.1700

Data Index: 1455551 (Python Code)

Max Activating Token Index: 274

Text #7

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.9070. Min Act: -0.1700

Data Index: 1512653 (Python Code)

Max Activating Token Index: 478

Text #8

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.9070. Min Act: -0.1700

Data Index: 1413466 (Python Code)

Max Activating Token Index: 800

Text #9

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.8757. Min Act: -0.1700

Data Index: 96111 (C4 (Web Text))

Max Activating Token Index: 198

Text #10

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.8757. Min Act: -0.1700

Data Index: 1385620 (Python Code)

Max Activating Token Index: 190

Text #11

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.9695. Min Act: -0.1700

Data Index: 132485 (C4 (Web Text))

Max Activating Token Index: 229

Text #12

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.7507. Min Act: -0.1700

Data Index: 1205113 (C4 (Web Text))

Max Activating Token Index: 500

Text #13

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.7820. Min Act: -0.1700

Data Index: 817258 (C4 (Web Text))

Max Activating Token Index: 268

Text #14

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.8132. Min Act: -0.1700

Data Index: 1488170 (Python Code)

Max Activating Token Index: 726

Text #15

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.8132. Min Act: -0.1700

Data Index: 1524818 (Python Code)

Max Activating Token Index: 93

Text #16

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.8445. Min Act: -0.1700

Data Index: 385978 (C4 (Web Text))

Max Activating Token Index: 156

Text #17

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.8445. Min Act: -0.1700

Data Index: 1019226 (C4 (Web Text))

Max Activating Token Index: 329

Text #18

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.8445. Min Act: -0.1700

Data Index: 97488 (C4 (Web Text))

Max Activating Token Index: 264

Text #19

Max Range: 5.0007. Min Range: -5.0007

Max Act: 4.8757. Min Act: -0.1700

Data Index: 1163844 (C4 (Web Text))

Max Activating Token Index: 361