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

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

Neuron 1997 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3515. Min Act: -0.1700

Data Index: 206741 (C4 (Web Text))

Max Activating Token Index: 521

Text #1

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3358. Min Act: -0.1700

Data Index: 1080977 (C4 (Web Text))

Max Activating Token Index: 362

Text #2

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3358. Min Act: -0.1700

Data Index: 145787 (C4 (Web Text))

Max Activating Token Index: 544

Text #3

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3515. Min Act: -0.1700

Data Index: 985232 (C4 (Web Text))

Max Activating Token Index: 178

Text #4

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3358. Min Act: -0.1700

Data Index: 913956 (C4 (Web Text))

Max Activating Token Index: 32

Text #5

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3201. Min Act: -0.1700

Data Index: 1101240 (C4 (Web Text))

Max Activating Token Index: 647

Text #6

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2887. Min Act: -0.1700

Data Index: 697555 (C4 (Web Text))

Max Activating Token Index: 346

Text #7

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3044. Min Act: -0.1700

Data Index: 287002 (C4 (Web Text))

Max Activating Token Index: 432

Text #8

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2887. Min Act: -0.1700

Data Index: 789705 (C4 (Web Text))

Max Activating Token Index: 216

Text #9

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3044. Min Act: -0.1700

Data Index: 381068 (C4 (Web Text))

Max Activating Token Index: 146

Text #10

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3044. Min Act: -0.1700

Data Index: 34270 (C4 (Web Text))

Max Activating Token Index: 709

Text #11

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.3044. Min Act: -0.1700

Data Index: 356746 (C4 (Web Text))

Max Activating Token Index: 48

Text #12

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2730. Min Act: -0.1700

Data Index: 1145772 (C4 (Web Text))

Max Activating Token Index: 645

Text #13

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2730. Min Act: -0.1700

Data Index: 1169364 (C4 (Web Text))

Max Activating Token Index: 221

Text #14

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2887. Min Act: -0.1700

Data Index: 167327 (C4 (Web Text))

Max Activating Token Index: 687

Text #15

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2887. Min Act: -0.1700

Data Index: 465079 (C4 (Web Text))

Max Activating Token Index: 398

Text #16

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2573. Min Act: -0.1700

Data Index: 529102 (C4 (Web Text))

Max Activating Token Index: 692

Text #17

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2416. Min Act: -0.1700

Data Index: 310209 (C4 (Web Text))

Max Activating Token Index: 779

Text #18

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2416. Min Act: -0.1700

Data Index: 11103 (C4 (Web Text))

Max Activating Token Index: 375

Text #19

Max Range: 3.3515. Min Range: -3.3515

Max Act: 3.2573. Min Act: -0.1700

Data Index: 1045179 (C4 (Web Text))

Max Activating Token Index: 568