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

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

Neuron 164 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.2969. Min Range: -3.2969

Max Act: 3.2969. Min Act: -0.1700

Data Index: 1518936 (Python Code)

Max Activating Token Index: 925

Text #1

Max Range: 3.2969. Min Range: -3.2969

Max Act: 3.2655. Min Act: -0.1700

Data Index: 1556657 (Python Code)

Max Activating Token Index: 61

Text #2

Max Range: 3.2969. Min Range: -3.2969

Max Act: 3.1868. Min Act: -0.1700

Data Index: 1539364 (Python Code)

Max Activating Token Index: 796

Text #3

Max Range: 3.2969. Min Range: -3.2969

Max Act: 3.1081. Min Act: -0.1700

Data Index: 1557591 (Python Code)

Max Activating Token Index: 674

Text #4

Max Range: 3.2969. Min Range: -3.2969

Max Act: 3.0608. Min Act: -0.1700

Data Index: 1472801 (Python Code)

Max Activating Token Index: 1014

Text #5

Max Range: 3.2969. Min Range: -3.2969

Max Act: 3.0923. Min Act: -0.1700

Data Index: 1376867 (Python Code)

Max Activating Token Index: 261

Text #6

Max Range: 3.2969. Min Range: -3.2969

Max Act: 3.0292. Min Act: -0.1700

Data Index: 1588007 (Python Code)

Max Activating Token Index: 998

Text #7

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.9976. Min Act: -0.1700

Data Index: 1497913 (Python Code)

Max Activating Token Index: 715

Text #8

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.9818. Min Act: -0.1700

Data Index: 1370742 (Python Code)

Max Activating Token Index: 916

Text #9

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.9659. Min Act: -0.1700

Data Index: 1514409 (Python Code)

Max Activating Token Index: 975

Text #10

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.9501. Min Act: -0.1700

Data Index: 1594920 (Python Code)

Max Activating Token Index: 853

Text #11

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.9026. Min Act: -0.1700

Data Index: 1566991 (Python Code)

Max Activating Token Index: 192

Text #12

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.8708. Min Act: -0.1700

Data Index: 1618865 (Python Code)

Max Activating Token Index: 1022

Text #13

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.9184. Min Act: -0.1700

Data Index: 413861 (C4 (Web Text))

Max Activating Token Index: 800

Text #14

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.8708. Min Act: -0.1700

Data Index: 1455840 (Python Code)

Max Activating Token Index: 1001

Text #15

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.8708. Min Act: -0.1700

Data Index: 1477275 (Python Code)

Max Activating Token Index: 203

Text #16

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.8867. Min Act: -0.1700

Data Index: 1465298 (Python Code)

Max Activating Token Index: 787

Text #17

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.8390. Min Act: -0.1700

Data Index: 1511338 (Python Code)

Max Activating Token Index: 386

Text #18

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.8231. Min Act: -0.1700

Data Index: 1427421 (Python Code)

Max Activating Token Index: 721

Text #19

Max Range: 3.2969. Min Range: -3.2969

Max Act: 2.8549. Min Act: -0.1700

Data Index: 1561571 (Python Code)

Max Activating Token Index: 999