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

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

Neuron 160 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: 5.4869. Min Range: -5.4869

Max Act: 5.4869. Min Act: -0.1700

Data Index: 289407 (C4 (Web Text))

Max Activating Token Index: 975

Text #1

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.3619. Min Act: -0.1700

Data Index: 117226 (C4 (Web Text))

Max Activating Token Index: 117

Text #2

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.3619. Min Act: -0.1700

Data Index: 440665 (C4 (Web Text))

Max Activating Token Index: 988

Text #3

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.3619. Min Act: -0.1700

Data Index: 995488 (C4 (Web Text))

Max Activating Token Index: 274

Text #4

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2681. Min Act: -0.1700

Data Index: 107612 (C4 (Web Text))

Max Activating Token Index: 601

Text #5

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2994. Min Act: -0.1700

Data Index: 616906 (C4 (Web Text))

Max Activating Token Index: 619

Text #6

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2681. Min Act: -0.1700

Data Index: 650829 (C4 (Web Text))

Max Activating Token Index: 869

Text #7

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2681. Min Act: -0.1700

Data Index: 34780 (C4 (Web Text))

Max Activating Token Index: 998

Text #8

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2994. Min Act: -0.1700

Data Index: 50842 (C4 (Web Text))

Max Activating Token Index: 866

Text #9

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2369. Min Act: -0.1700

Data Index: 608090 (C4 (Web Text))

Max Activating Token Index: 987

Text #10

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2369. Min Act: -0.1700

Data Index: 226485 (C4 (Web Text))

Max Activating Token Index: 915

Text #11

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2056. Min Act: -0.1700

Data Index: 837800 (C4 (Web Text))

Max Activating Token Index: 567

Text #12

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2369. Min Act: -0.1700

Data Index: 1197601 (C4 (Web Text))

Max Activating Token Index: 940

Text #13

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2369. Min Act: -0.1700

Data Index: 1291713 (C4 (Web Text))

Max Activating Token Index: 178

Text #14

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2369. Min Act: -0.1700

Data Index: 211168 (C4 (Web Text))

Max Activating Token Index: 415

Text #15

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2681. Min Act: -0.1700

Data Index: 1354390 (C4 (Web Text))

Max Activating Token Index: 700

Text #16

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2056. Min Act: -0.1700

Data Index: 142075 (C4 (Web Text))

Max Activating Token Index: 162

Text #17

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2056. Min Act: -0.1700

Data Index: 795452 (C4 (Web Text))

Max Activating Token Index: 618

Text #18

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.1119. Min Act: -0.1700

Data Index: 550673 (C4 (Web Text))

Max Activating Token Index: 312

Text #19

Max Range: 5.4869. Min Range: -5.4869

Max Act: 5.2056. Min Act: -0.1700

Data Index: 385640 (C4 (Web Text))

Max Activating Token Index: 97