Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1214 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 4.6771. Min Range: -4.6771

Max Act: 4.6771. Min Act: -0.1700

Data Index: 268408 (C4 (Web Text))

Max Activating Token Index: 553

Text #1

Max Range: 4.6771. Min Range: -4.6771

Max Act: 4.3021. Min Act: -0.1700

Data Index: 113011 (C4 (Web Text))

Max Activating Token Index: 839

Text #2

Max Range: 4.6771. Min Range: -4.6771

Max Act: 4.2396. Min Act: -0.1700

Data Index: 1377734 (Python Code)

Max Activating Token Index: 744

Text #3

Max Range: 4.6771. Min Range: -4.6771

Max Act: 4.1458. Min Act: -0.1700

Data Index: 1399846 (Python Code)

Max Activating Token Index: 603

Text #4

Max Range: 4.6771. Min Range: -4.6771

Max Act: 4.0520. Min Act: -0.1700

Data Index: 1114603 (C4 (Web Text))

Max Activating Token Index: 430

Text #5

Max Range: 4.6771. Min Range: -4.6771

Max Act: 4.0208. Min Act: -0.1700

Data Index: 196977 (C4 (Web Text))

Max Activating Token Index: 479

Text #6

Max Range: 4.6771. Min Range: -4.6771

Max Act: 4.0520. Min Act: -0.1700

Data Index: 327572 (C4 (Web Text))

Max Activating Token Index: 799

Text #7

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9895. Min Act: -0.1700

Data Index: 1470508 (Python Code)

Max Activating Token Index: 753

Text #8

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9895. Min Act: -0.1700

Data Index: 77325 (C4 (Web Text))

Max Activating Token Index: 524

Text #9

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9895. Min Act: -0.1700

Data Index: 614757 (C4 (Web Text))

Max Activating Token Index: 1019

Text #10

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9270. Min Act: -0.1700

Data Index: 327242 (C4 (Web Text))

Max Activating Token Index: 861

Text #11

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9270. Min Act: -0.1700

Data Index: 426248 (C4 (Web Text))

Max Activating Token Index: 316

Text #12

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9270. Min Act: -0.1700

Data Index: 229843 (C4 (Web Text))

Max Activating Token Index: 958

Text #13

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9582. Min Act: -0.1700

Data Index: 1035891 (C4 (Web Text))

Max Activating Token Index: 460

Text #14

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9270. Min Act: -0.1700

Data Index: 1640438 (Python Code)

Max Activating Token Index: 487

Text #15

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9582. Min Act: -0.1700

Data Index: 71473 (C4 (Web Text))

Max Activating Token Index: 981

Text #16

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.9270. Min Act: -0.1700

Data Index: 1022128 (C4 (Web Text))

Max Activating Token Index: 715

Text #17

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.8957. Min Act: -0.1700

Data Index: 1330048 (C4 (Web Text))

Max Activating Token Index: 805

Text #18

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.8644. Min Act: -0.1700

Data Index: 1180707 (C4 (Web Text))

Max Activating Token Index: 377

Text #19

Max Range: 4.6771. Min Range: -4.6771

Max Act: 3.8957. Min Act: -0.1700

Data Index: 226279 (C4 (Web Text))

Max Activating Token Index: 836