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

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

Neuron 44 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: 5.5371. Min Range: -5.5371

Max Act: 5.5371. Min Act: -0.1700

Data Index: 120300 (C4 (Web Text))

Max Activating Token Index: 23

Text #1

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.4121. Min Act: -0.1700

Data Index: 248449 (C4 (Web Text))

Max Activating Token Index: 839

Text #2

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.3809. Min Act: -0.1700

Data Index: 195218 (C4 (Web Text))

Max Activating Token Index: 1008

Text #3

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.3184. Min Act: -0.1700

Data Index: 1199705 (C4 (Web Text))

Max Activating Token Index: 693

Text #4

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.3184. Min Act: -0.1700

Data Index: 529388 (C4 (Web Text))

Max Activating Token Index: 1016

Text #5

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.3184. Min Act: -0.1700

Data Index: 995641 (C4 (Web Text))

Max Activating Token Index: 639

Text #6

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.2559. Min Act: -0.1700

Data Index: 392863 (C4 (Web Text))

Max Activating Token Index: 967

Text #7

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.2559. Min Act: -0.1700

Data Index: 908211 (C4 (Web Text))

Max Activating Token Index: 884

Text #8

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.2871. Min Act: -0.1700

Data Index: 364968 (C4 (Web Text))

Max Activating Token Index: 606

Text #9

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.2559. Min Act: -0.1700

Data Index: 634939 (C4 (Web Text))

Max Activating Token Index: 216

Text #10

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.2246. Min Act: -0.1699

Data Index: 396908 (C4 (Web Text))

Max Activating Token Index: 217

Text #11

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.2246. Min Act: -0.1700

Data Index: 859648 (C4 (Web Text))

Max Activating Token Index: 164

Text #12

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.2559. Min Act: -0.1700

Data Index: 1064192 (C4 (Web Text))

Max Activating Token Index: 860

Text #13

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.2246. Min Act: -0.1700

Data Index: 1222925 (C4 (Web Text))

Max Activating Token Index: 52

Text #14

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.1934. Min Act: -0.1700

Data Index: 311574 (C4 (Web Text))

Max Activating Token Index: 226

Text #15

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.1934. Min Act: -0.1700

Data Index: 238652 (C4 (Web Text))

Max Activating Token Index: 719

Text #16

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.1621. Min Act: -0.1700

Data Index: 418103 (C4 (Web Text))

Max Activating Token Index: 894

Text #17

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.1621. Min Act: -0.1700

Data Index: 629232 (C4 (Web Text))

Max Activating Token Index: 548

Text #18

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.1934. Min Act: -0.1700

Data Index: 298006 (C4 (Web Text))

Max Activating Token Index: 679

Text #19

Max Range: 5.5371. Min Range: -5.5371

Max Act: 5.1621. Min Act: -0.1700

Data Index: 702176 (C4 (Web Text))

Max Activating Token Index: 390