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

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

Neuron 16 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: 4.0551. Min Range: -4.0551

Max Act: 4.0551. Min Act: -0.1700

Data Index: 132166 (C4 (Web Text))

Max Activating Token Index: 9

Text #1

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.9926. Min Act: -0.1700

Data Index: 1050874 (C4 (Web Text))

Max Activating Token Index: 982

Text #2

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.9613. Min Act: -0.1700

Data Index: 1278500 (C4 (Web Text))

Max Activating Token Index: 134

Text #3

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.9301. Min Act: -0.1700

Data Index: 857734 (C4 (Web Text))

Max Activating Token Index: 186

Text #4

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8988. Min Act: -0.1700

Data Index: 849347 (C4 (Web Text))

Max Activating Token Index: 882

Text #5

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.9301. Min Act: -0.1700

Data Index: 407111 (C4 (Web Text))

Max Activating Token Index: 838

Text #6

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.9301. Min Act: -0.1700

Data Index: 261874 (C4 (Web Text))

Max Activating Token Index: 976

Text #7

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8988. Min Act: -0.1700

Data Index: 1264088 (C4 (Web Text))

Max Activating Token Index: 287

Text #8

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8988. Min Act: -0.1700

Data Index: 1190192 (C4 (Web Text))

Max Activating Token Index: 845

Text #9

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8988. Min Act: -0.1700

Data Index: 453577 (C4 (Web Text))

Max Activating Token Index: 742

Text #10

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.9301. Min Act: -0.1700

Data Index: 1353381 (C4 (Web Text))

Max Activating Token Index: 483

Text #11

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8988. Min Act: -0.1700

Data Index: 717774 (C4 (Web Text))

Max Activating Token Index: 815

Text #12

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8363. Min Act: -0.1700

Data Index: 1141816 (C4 (Web Text))

Max Activating Token Index: 842

Text #13

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8363. Min Act: -0.1700

Data Index: 218335 (C4 (Web Text))

Max Activating Token Index: 566

Text #14

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8675. Min Act: -0.1700

Data Index: 804056 (C4 (Web Text))

Max Activating Token Index: 526

Text #15

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8988. Min Act: -0.1700

Data Index: 818041 (C4 (Web Text))

Max Activating Token Index: 823

Text #16

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8675. Min Act: -0.1700

Data Index: 1222482 (C4 (Web Text))

Max Activating Token Index: 900

Text #17

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8363. Min Act: -0.1700

Data Index: 483242 (C4 (Web Text))

Max Activating Token Index: 630

Text #18

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8363. Min Act: -0.1700

Data Index: 503919 (C4 (Web Text))

Max Activating Token Index: 91

Text #19

Max Range: 4.0551. Min Range: -4.0551

Max Act: 3.8363. Min Act: -0.1700

Data Index: 1174783 (C4 (Web Text))

Max Activating Token Index: 829