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

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

Neuron 1490 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: 5.4189. Min Range: -5.4189

Max Act: 5.4189. Min Act: -0.1700

Data Index: 384262 (C4 (Web Text))

Max Activating Token Index: 447

Text #1

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.3251. Min Act: -0.1700

Data Index: 1010541 (C4 (Web Text))

Max Activating Token Index: 142

Text #2

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.2939. Min Act: -0.1700

Data Index: 943495 (C4 (Web Text))

Max Activating Token Index: 715

Text #3

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.3564. Min Act: -0.1700

Data Index: 364108 (C4 (Web Text))

Max Activating Token Index: 877

Text #4

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.2626. Min Act: -0.1700

Data Index: 256029 (C4 (Web Text))

Max Activating Token Index: 123

Text #5

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.2626. Min Act: -0.1700

Data Index: 189324 (C4 (Web Text))

Max Activating Token Index: 339

Text #6

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.2314. Min Act: -0.1700

Data Index: 911511 (C4 (Web Text))

Max Activating Token Index: 128

Text #7

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.2001. Min Act: -0.1700

Data Index: 1119840 (C4 (Web Text))

Max Activating Token Index: 829

Text #8

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.2001. Min Act: -0.1700

Data Index: 507479 (C4 (Web Text))

Max Activating Token Index: 95

Text #9

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.0907. Min Act: -0.1700

Data Index: 556984 (C4 (Web Text))

Max Activating Token Index: 1006

Text #10

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.1220. Min Act: -0.1700

Data Index: 624269 (C4 (Web Text))

Max Activating Token Index: 751

Text #11

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.1220. Min Act: -0.1700

Data Index: 240263 (C4 (Web Text))

Max Activating Token Index: 223

Text #12

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.0751. Min Act: -0.1700

Data Index: 17858 (C4 (Web Text))

Max Activating Token Index: 379

Text #13

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.1376. Min Act: -0.1700

Data Index: 326244 (C4 (Web Text))

Max Activating Token Index: 346

Text #14

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.0751. Min Act: -0.1700

Data Index: 1000346 (C4 (Web Text))

Max Activating Token Index: 226

Text #15

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.1064. Min Act: -0.1700

Data Index: 1011534 (C4 (Web Text))

Max Activating Token Index: 119

Text #16

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.1064. Min Act: -0.1700

Data Index: 551142 (C4 (Web Text))

Max Activating Token Index: 497

Text #17

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.1064. Min Act: -0.1700

Data Index: 1104282 (C4 (Web Text))

Max Activating Token Index: 640

Text #18

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.0439. Min Act: -0.1700

Data Index: 671548 (C4 (Web Text))

Max Activating Token Index: 227

Text #19

Max Range: 5.4189. Min Range: -5.4189

Max Act: 5.0751. Min Act: -0.1699

Data Index: 164670 (C4 (Web Text))

Max Activating Token Index: 596