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

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

Neuron 831 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: 4.9690. Min Range: -4.9690

Max Act: 4.9221. Min Act: -0.1700

Data Index: 1058551 (C4 (Web Text))

Max Activating Token Index: 280

Text #1

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.9690. Min Act: -0.1700

Data Index: 180020 (C4 (Web Text))

Max Activating Token Index: 359

Text #2

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.9378. Min Act: -0.1700

Data Index: 1200140 (C4 (Web Text))

Max Activating Token Index: 715

Text #3

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.8128. Min Act: -0.1699

Data Index: 372800 (C4 (Web Text))

Max Activating Token Index: 710

Text #4

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.8440. Min Act: -0.1700

Data Index: 804504 (C4 (Web Text))

Max Activating Token Index: 262

Text #5

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.8128. Min Act: -0.1700

Data Index: 329027 (C4 (Web Text))

Max Activating Token Index: 437

Text #6

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.7815. Min Act: -0.1700

Data Index: 291359 (C4 (Web Text))

Max Activating Token Index: 507

Text #7

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.7346. Min Act: -0.1700

Data Index: 944403 (C4 (Web Text))

Max Activating Token Index: 171

Text #8

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.7503. Min Act: -0.1700

Data Index: 296036 (C4 (Web Text))

Max Activating Token Index: 877

Text #9

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.7034. Min Act: -0.1700

Data Index: 987238 (C4 (Web Text))

Max Activating Token Index: 599

Text #10

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.6721. Min Act: -0.1700

Data Index: 712202 (C4 (Web Text))

Max Activating Token Index: 792

Text #11

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.6096. Min Act: -0.1700

Data Index: 1200573 (C4 (Web Text))

Max Activating Token Index: 261

Text #12

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.6409. Min Act: -0.1700

Data Index: 471381 (C4 (Web Text))

Max Activating Token Index: 156

Text #13

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.6565. Min Act: -0.1700

Data Index: 566497 (C4 (Web Text))

Max Activating Token Index: 807

Text #14

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.6253. Min Act: -0.1700

Data Index: 725628 (C4 (Web Text))

Max Activating Token Index: 458

Text #15

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.6253. Min Act: -0.1700

Data Index: 1048911 (C4 (Web Text))

Max Activating Token Index: 732

Text #16

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.5940. Min Act: -0.1700

Data Index: 393084 (C4 (Web Text))

Max Activating Token Index: 323

Text #17

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.6565. Min Act: -0.1700

Data Index: 565459 (C4 (Web Text))

Max Activating Token Index: 210

Text #18

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.5003. Min Act: -0.1700

Data Index: 629621 (C4 (Web Text))

Max Activating Token Index: 329

Text #19

Max Range: 4.9690. Min Range: -4.9690

Max Act: 4.6253. Min Act: -0.1700

Data Index: 47133 (C4 (Web Text))

Max Activating Token Index: 21