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

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

Neuron 1950 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: 7.4888. Min Range: -7.4888

Max Act: 7.4888. Min Act: -0.1700

Data Index: 924907 (C4 (Web Text))

Max Activating Token Index: 430

Text #1

Max Range: 7.4888. Min Range: -7.4888

Max Act: 7.2076. Min Act: -0.1700

Data Index: 792603 (C4 (Web Text))

Max Activating Token Index: 153

Text #2

Max Range: 7.4888. Min Range: -7.4888

Max Act: 7.2076. Min Act: -0.1700

Data Index: 3223 (C4 (Web Text))

Max Activating Token Index: 251

Text #3

Max Range: 7.4888. Min Range: -7.4888

Max Act: 7.2388. Min Act: -0.1700

Data Index: 352595 (C4 (Web Text))

Max Activating Token Index: 259

Text #4

Max Range: 7.4888. Min Range: -7.4888

Max Act: 7.0513. Min Act: -0.1700

Data Index: 1243958 (C4 (Web Text))

Max Activating Token Index: 145

Text #5

Max Range: 7.4888. Min Range: -7.4888

Max Act: 7.1138. Min Act: -0.1700

Data Index: 980651 (C4 (Web Text))

Max Activating Token Index: 610

Text #6

Max Range: 7.4888. Min Range: -7.4888

Max Act: 7.0513. Min Act: -0.1700

Data Index: 82926 (C4 (Web Text))

Max Activating Token Index: 1004

Text #7

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.9263. Min Act: -0.1700

Data Index: 913612 (C4 (Web Text))

Max Activating Token Index: 196

Text #8

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.9263. Min Act: -0.1700

Data Index: 326950 (C4 (Web Text))

Max Activating Token Index: 1018

Text #9

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.9263. Min Act: -0.1700

Data Index: 392926 (C4 (Web Text))

Max Activating Token Index: 569

Text #10

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.9263. Min Act: -0.1700

Data Index: 665312 (C4 (Web Text))

Max Activating Token Index: 167

Text #11

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8951. Min Act: -0.1700

Data Index: 876535 (C4 (Web Text))

Max Activating Token Index: 284

Text #12

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8638. Min Act: -0.1699

Data Index: 43554 (C4 (Web Text))

Max Activating Token Index: 445

Text #13

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8638. Min Act: -0.1700

Data Index: 21522 (C4 (Web Text))

Max Activating Token Index: 231

Text #14

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8326. Min Act: -0.1700

Data Index: 628749 (C4 (Web Text))

Max Activating Token Index: 45

Text #15

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8951. Min Act: -0.1700

Data Index: 557187 (C4 (Web Text))

Max Activating Token Index: 439

Text #16

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8013. Min Act: -0.1700

Data Index: 242727 (C4 (Web Text))

Max Activating Token Index: 852

Text #17

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8326. Min Act: -0.1699

Data Index: 680918 (C4 (Web Text))

Max Activating Token Index: 753

Text #18

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8326. Min Act: -0.1700

Data Index: 1082554 (C4 (Web Text))

Max Activating Token Index: 854

Text #19

Max Range: 7.4888. Min Range: -7.4888

Max Act: 6.8013. Min Act: -0.1700

Data Index: 1162910 (C4 (Web Text))

Max Activating Token Index: 446