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

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

Neuron 662 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: 5.8870. Min Range: -5.8870

Max Act: 5.8870. Min Act: -0.1700

Data Index: 174094 (C4 (Web Text))

Max Activating Token Index: 425

Text #1

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.8558. Min Act: -0.1700

Data Index: 483921 (C4 (Web Text))

Max Activating Token Index: 258

Text #2

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.6370. Min Act: -0.1700

Data Index: 304666 (C4 (Web Text))

Max Activating Token Index: 384

Text #3

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.5745. Min Act: -0.1700

Data Index: 291036 (C4 (Web Text))

Max Activating Token Index: 899

Text #4

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.6058. Min Act: -0.1700

Data Index: 822765 (C4 (Web Text))

Max Activating Token Index: 648

Text #5

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.5433. Min Act: -0.1700

Data Index: 1031120 (C4 (Web Text))

Max Activating Token Index: 687

Text #6

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.5433. Min Act: -0.1700

Data Index: 1142134 (C4 (Web Text))

Max Activating Token Index: 773

Text #7

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.5433. Min Act: -0.1700

Data Index: 391320 (C4 (Web Text))

Max Activating Token Index: 702

Text #8

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4808. Min Act: -0.1700

Data Index: 479239 (C4 (Web Text))

Max Activating Token Index: 578

Text #9

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4808. Min Act: -0.1700

Data Index: 1011142 (C4 (Web Text))

Max Activating Token Index: 346

Text #10

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4808. Min Act: -0.1700

Data Index: 435405 (C4 (Web Text))

Max Activating Token Index: 542

Text #11

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4495. Min Act: -0.1700

Data Index: 969044 (C4 (Web Text))

Max Activating Token Index: 313

Text #12

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4183. Min Act: -0.1700

Data Index: 622868 (C4 (Web Text))

Max Activating Token Index: 107

Text #13

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4808. Min Act: -0.1700

Data Index: 1239541 (C4 (Web Text))

Max Activating Token Index: 522

Text #14

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4495. Min Act: -0.1700

Data Index: 40756 (C4 (Web Text))

Max Activating Token Index: 602

Text #15

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4183. Min Act: -0.1700

Data Index: 1327008 (C4 (Web Text))

Max Activating Token Index: 280

Text #16

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4495. Min Act: -0.1700

Data Index: 338036 (C4 (Web Text))

Max Activating Token Index: 527

Text #17

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.3870. Min Act: -0.1700

Data Index: 152375 (C4 (Web Text))

Max Activating Token Index: 966

Text #18

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.4183. Min Act: -0.1700

Data Index: 637895 (C4 (Web Text))

Max Activating Token Index: 982

Text #19

Max Range: 5.8870. Min Range: -5.8870

Max Act: 5.3870. Min Act: -0.1700

Data Index: 648777 (C4 (Web Text))

Max Activating Token Index: 875