Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2927 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 2.1971. Min Range: -2.1971

Max Act: 2.1971. Min Act: -0.0001

Data Index: 1623391 (Python Code)

Max Activating Token Index: 451

Text #1

Max Range: 2.1971. Min Range: -2.1971

Max Act: 2.0133. Min Act: -0.0001

Data Index: 943851 (C4 (Web Text))

Max Activating Token Index: 899

Text #2

Max Range: 2.1971. Min Range: -2.1971

Max Act: 2.0193. Min Act: -0.0001

Data Index: 936940 (C4 (Web Text))

Max Activating Token Index: 212

Text #3

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.9708. Min Act: -0.0001

Data Index: 1487094 (Python Code)

Max Activating Token Index: 960

Text #4

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.8540. Min Act: -0.0001

Data Index: 1626271 (Python Code)

Max Activating Token Index: 675

Text #5

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.8486. Min Act: -0.0001

Data Index: 1529730 (Python Code)

Max Activating Token Index: 603

Text #6

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.5465. Min Act: -0.0001

Data Index: 1546891 (Python Code)

Max Activating Token Index: 906

Text #7

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.8999. Min Act: -0.0001

Data Index: 474169 (C4 (Web Text))

Max Activating Token Index: 765

Text #8

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.6725. Min Act: -0.0001

Data Index: 116550 (C4 (Web Text))

Max Activating Token Index: 614

Text #9

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.6745. Min Act: -0.0001

Data Index: 435341 (C4 (Web Text))

Max Activating Token Index: 812

Text #10

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.7649. Min Act: -0.0001

Data Index: 678701 (C4 (Web Text))

Max Activating Token Index: 767

Text #11

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.6642. Min Act: -0.0001

Data Index: 1371982 (Python Code)

Max Activating Token Index: 100

Text #12

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.6596. Min Act: -0.0001

Data Index: 1633928 (Python Code)

Max Activating Token Index: 585

Text #13

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.7446. Min Act: -0.0001

Data Index: 1376710 (Python Code)

Max Activating Token Index: 141

Text #14

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.6409. Min Act: -0.0001

Data Index: 696596 (C4 (Web Text))

Max Activating Token Index: 132

Text #15

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.8301. Min Act: -0.0001

Data Index: 462269 (C4 (Web Text))

Max Activating Token Index: 648

Text #16

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.8259. Min Act: -0.0001

Data Index: 947883 (C4 (Web Text))

Max Activating Token Index: 731

Text #17

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.5976. Min Act: -0.0001

Data Index: 452475 (C4 (Web Text))

Max Activating Token Index: 1020

Text #18

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.5976. Min Act: -0.0001

Data Index: 351806 (C4 (Web Text))

Max Activating Token Index: 18

Text #19

Max Range: 2.1971. Min Range: -2.1971

Max Act: 1.4820. Min Act: -0.0001

Data Index: 549904 (C4 (Web Text))

Max Activating Token Index: 743