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

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

Neuron 1216 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.4456. Min Act: -0.0001

Data Index: 819411 (C4 (Web Text))

Max Activating Token Index: 752

Text #1

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.4173. Min Act: -0.0001

Data Index: 1529103 (Python Code)

Max Activating Token Index: 855

Text #2

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.4297. Min Act: -0.0001

Data Index: 888840 (C4 (Web Text))

Max Activating Token Index: 723

Text #3

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.4426. Min Act: -0.0000

Data Index: 1245425 (C4 (Web Text))

Max Activating Token Index: 182

Text #4

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3932. Min Act: -0.0001

Data Index: 448525 (C4 (Web Text))

Max Activating Token Index: 777

Text #5

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3659. Min Act: -0.0001

Data Index: 1386745 (Python Code)

Max Activating Token Index: 969

Text #6

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3413. Min Act: -0.0001

Data Index: 181520 (C4 (Web Text))

Max Activating Token Index: 165

Text #7

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3435. Min Act: -0.0001

Data Index: 1292277 (C4 (Web Text))

Max Activating Token Index: 931

Text #8

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3224. Min Act: -0.0000

Data Index: 383707 (C4 (Web Text))

Max Activating Token Index: 53

Text #9

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3775. Min Act: -0.0000

Data Index: 1071490 (C4 (Web Text))

Max Activating Token Index: 312

Text #10

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3193. Min Act: -0.0001

Data Index: 33377 (C4 (Web Text))

Max Activating Token Index: 907

Text #11

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.2950. Min Act: -0.0001

Data Index: 1538372 (Python Code)

Max Activating Token Index: 615

Text #12

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3055. Min Act: -0.0001

Data Index: 1469889 (Python Code)

Max Activating Token Index: 575

Text #13

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.2954. Min Act: -0.0001

Data Index: 529457 (C4 (Web Text))

Max Activating Token Index: 578

Text #14

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3043. Min Act: -0.0001

Data Index: 1501123 (Python Code)

Max Activating Token Index: 925

Text #15

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3120. Min Act: -0.0001

Data Index: 1541478 (Python Code)

Max Activating Token Index: 758

Text #16

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.2952. Min Act: -0.0001

Data Index: 466263 (C4 (Web Text))

Max Activating Token Index: 595

Text #17

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.2817. Min Act: -0.0000

Data Index: 816826 (C4 (Web Text))

Max Activating Token Index: 490

Text #18

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3002. Min Act: -0.0001

Data Index: 914013 (C4 (Web Text))

Max Activating Token Index: 853

Text #19

Max Range: 0.4456. Min Range: -0.4456

Max Act: 0.3127. Min Act: -0.0001

Data Index: 61828 (C4 (Web Text))

Max Activating Token Index: 164