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

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

Neuron 285 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: 5.4943. Min Range: -5.4943

Max Act: 5.4943. Min Act: -0.1700

Data Index: 703368 (C4 (Web Text))

Max Activating Token Index: 864

Text #1

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.3068. Min Act: -0.1700

Data Index: 1119612 (C4 (Web Text))

Max Activating Token Index: 521

Text #2

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.2130. Min Act: -0.1700

Data Index: 279196 (C4 (Web Text))

Max Activating Token Index: 656

Text #3

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.3068. Min Act: -0.1700

Data Index: 998220 (C4 (Web Text))

Max Activating Token Index: 613

Text #4

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.3068. Min Act: -0.1700

Data Index: 1110765 (C4 (Web Text))

Max Activating Token Index: 533

Text #5

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.2755. Min Act: -0.1700

Data Index: 1309371 (C4 (Web Text))

Max Activating Token Index: 987

Text #6

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.2755. Min Act: -0.1700

Data Index: 574372 (C4 (Web Text))

Max Activating Token Index: 253

Text #7

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.2755. Min Act: -0.1700

Data Index: 325961 (C4 (Web Text))

Max Activating Token Index: 981

Text #8

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.2443. Min Act: -0.1700

Data Index: 1283800 (C4 (Web Text))

Max Activating Token Index: 315

Text #9

Max Range: 5.4943. Min Range: -5.4943

Max Act: 4.9630. Min Act: -0.1700

Data Index: 10490 (C4 (Web Text))

Max Activating Token Index: 105

Text #10

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.0255. Min Act: -0.1700

Data Index: 1036000 (C4 (Web Text))

Max Activating Token Index: 548

Text #11

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.1818. Min Act: -0.1700

Data Index: 13342 (C4 (Web Text))

Max Activating Token Index: 512

Text #12

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.1818. Min Act: -0.1700

Data Index: 482842 (C4 (Web Text))

Max Activating Token Index: 642

Text #13

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.1505. Min Act: -0.1700

Data Index: 720541 (C4 (Web Text))

Max Activating Token Index: 726

Text #14

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.1193. Min Act: -0.1700

Data Index: 476003 (C4 (Web Text))

Max Activating Token Index: 466

Text #15

Max Range: 5.4943. Min Range: -5.4943

Max Act: 4.9943. Min Act: -0.1700

Data Index: 722608 (C4 (Web Text))

Max Activating Token Index: 364

Text #16

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.0880. Min Act: -0.1700

Data Index: 1122450 (C4 (Web Text))

Max Activating Token Index: 558

Text #17

Max Range: 5.4943. Min Range: -5.4943

Max Act: 4.9630. Min Act: -0.1700

Data Index: 273512 (C4 (Web Text))

Max Activating Token Index: 896

Text #18

Max Range: 5.4943. Min Range: -5.4943

Max Act: 4.9318. Min Act: -0.1700

Data Index: 998629 (C4 (Web Text))

Max Activating Token Index: 379

Text #19

Max Range: 5.4943. Min Range: -5.4943

Max Act: 5.0568. Min Act: -0.1700

Data Index: 966926 (C4 (Web Text))

Max Activating Token Index: 737