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

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

Neuron 253 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.2515. Min Act: -0.1700

Data Index: 552211 (C4 (Web Text))

Max Activating Token Index: 86

Text #1

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.2515. Min Act: -0.1700

Data Index: 1605881 (Python Code)

Max Activating Token Index: 215

Text #2

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.1571. Min Act: -0.1700

Data Index: 53931 (C4 (Web Text))

Max Activating Token Index: 144

Text #3

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.1256. Min Act: -0.1700

Data Index: 1283540 (C4 (Web Text))

Max Activating Token Index: 520

Text #4

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.1099. Min Act: -0.1700

Data Index: 1339535 (C4 (Web Text))

Max Activating Token Index: 200

Text #5

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.1256. Min Act: -0.1700

Data Index: 1336205 (C4 (Web Text))

Max Activating Token Index: 454

Text #6

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.1099. Min Act: -0.1700

Data Index: 1028543 (C4 (Web Text))

Max Activating Token Index: 437

Text #7

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.1256. Min Act: -0.1700

Data Index: 585908 (C4 (Web Text))

Max Activating Token Index: 797

Text #8

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0941. Min Act: -0.1700

Data Index: 259007 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.1099. Min Act: -0.1700

Data Index: 310232 (C4 (Web Text))

Max Activating Token Index: 648

Text #10

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0783. Min Act: -0.1700

Data Index: 147113 (C4 (Web Text))

Max Activating Token Index: 701

Text #11

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0783. Min Act: -0.1700

Data Index: 858908 (C4 (Web Text))

Max Activating Token Index: 608

Text #12

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0783. Min Act: -0.1700

Data Index: 1304726 (C4 (Web Text))

Max Activating Token Index: 453

Text #13

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0941. Min Act: -0.1700

Data Index: 919313 (C4 (Web Text))

Max Activating Token Index: 642

Text #14

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.1256. Min Act: -0.1700

Data Index: 1603042 (Python Code)

Max Activating Token Index: 636

Text #15

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0783. Min Act: -0.1700

Data Index: 907826 (C4 (Web Text))

Max Activating Token Index: 256

Text #16

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0626. Min Act: -0.1700

Data Index: 893857 (C4 (Web Text))

Max Activating Token Index: 809

Text #17

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0783. Min Act: -0.1700

Data Index: 740420 (C4 (Web Text))

Max Activating Token Index: 187

Text #18

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0783. Min Act: -0.1700

Data Index: 22363 (C4 (Web Text))

Max Activating Token Index: 398

Text #19

Max Range: 3.2515. Min Range: -3.2515

Max Act: 3.0783. Min Act: -0.1700

Data Index: 162076 (C4 (Web Text))

Max Activating Token Index: 32