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

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

Neuron 1454 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: 5.4966. Min Range: -5.4966

Max Act: 5.4966. Min Act: -0.1700

Data Index: 1272502 (C4 (Web Text))

Max Activating Token Index: 459

Text #1

Max Range: 5.4966. Min Range: -5.4966

Max Act: 5.1216. Min Act: -0.1700

Data Index: 192570 (C4 (Web Text))

Max Activating Token Index: 935

Text #2

Max Range: 5.4966. Min Range: -5.4966

Max Act: 5.1216. Min Act: -0.1700

Data Index: 1115480 (C4 (Web Text))

Max Activating Token Index: 580

Text #3

Max Range: 5.4966. Min Range: -5.4966

Max Act: 5.0591. Min Act: -0.1700

Data Index: 49473 (C4 (Web Text))

Max Activating Token Index: 202

Text #4

Max Range: 5.4966. Min Range: -5.4966

Max Act: 5.0591. Min Act: -0.1700

Data Index: 562029 (C4 (Web Text))

Max Activating Token Index: 562

Text #5

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.9966. Min Act: -0.1700

Data Index: 732012 (C4 (Web Text))

Max Activating Token Index: 430

Text #6

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.9341. Min Act: -0.1700

Data Index: 615148 (C4 (Web Text))

Max Activating Token Index: 285

Text #7

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.9028. Min Act: -0.1700

Data Index: 1327165 (C4 (Web Text))

Max Activating Token Index: 425

Text #8

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.9028. Min Act: -0.1700

Data Index: 70008 (C4 (Web Text))

Max Activating Token Index: 150

Text #9

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.8716. Min Act: -0.1700

Data Index: 753498 (C4 (Web Text))

Max Activating Token Index: 388

Text #10

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.8403. Min Act: -0.1700

Data Index: 608186 (C4 (Web Text))

Max Activating Token Index: 402

Text #11

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.8091. Min Act: -0.1700

Data Index: 1338850 (C4 (Web Text))

Max Activating Token Index: 1022

Text #12

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.8091. Min Act: -0.1700

Data Index: 1144404 (C4 (Web Text))

Max Activating Token Index: 798

Text #13

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.8403. Min Act: -0.1700

Data Index: 1134056 (C4 (Web Text))

Max Activating Token Index: 508

Text #14

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.7153. Min Act: -0.1700

Data Index: 891331 (C4 (Web Text))

Max Activating Token Index: 838

Text #15

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.8091. Min Act: -0.1700

Data Index: 561755 (C4 (Web Text))

Max Activating Token Index: 678

Text #16

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.8091. Min Act: -0.1699

Data Index: 1268725 (C4 (Web Text))

Max Activating Token Index: 952

Text #17

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.7466. Min Act: -0.1700

Data Index: 657414 (C4 (Web Text))

Max Activating Token Index: 849

Text #18

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.7466. Min Act: -0.1700

Data Index: 363721 (C4 (Web Text))

Max Activating Token Index: 357

Text #19

Max Range: 5.4966. Min Range: -5.4966

Max Act: 4.7466. Min Act: -0.1700

Data Index: 409865 (C4 (Web Text))

Max Activating Token Index: 572