Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 770 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.0464. Min Range: -2.0464

Max Act: 2.0464. Min Act: -0.0000

Data Index: 1626412 (Python Code)

Max Activating Token Index: 861

Text #1

Max Range: 2.0464. Min Range: -2.0464

Max Act: 2.0139. Min Act: -0.0000

Data Index: 1586828 (Python Code)

Max Activating Token Index: 28

Text #2

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.8792. Min Act: -0.0000

Data Index: 431251 (C4 (Web Text))

Max Activating Token Index: 236

Text #3

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.8495. Min Act: -0.0000

Data Index: 1372251 (Python Code)

Max Activating Token Index: 762

Text #4

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.7508. Min Act: -0.0000

Data Index: 1503137 (Python Code)

Max Activating Token Index: 607

Text #5

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.7828. Min Act: -0.0000

Data Index: 1440170 (Python Code)

Max Activating Token Index: 533

Text #6

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.7783. Min Act: -0.0000

Data Index: 1255283 (C4 (Web Text))

Max Activating Token Index: 993

Text #7

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.8678. Min Act: -0.0000

Data Index: 1444126 (Python Code)

Max Activating Token Index: 107

Text #8

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.7497. Min Act: -0.0000

Data Index: 1102251 (C4 (Web Text))

Max Activating Token Index: 977

Text #9

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.7555. Min Act: -0.0000

Data Index: 1526116 (Python Code)

Max Activating Token Index: 209

Text #10

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.6700. Min Act: -0.0000

Data Index: 894870 (C4 (Web Text))

Max Activating Token Index: 696

Text #11

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.6524. Min Act: -0.0000

Data Index: 1346795 (C4 (Web Text))

Max Activating Token Index: 275

Text #12

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.6628. Min Act: -0.0000

Data Index: 397650 (C4 (Web Text))

Max Activating Token Index: 286

Text #13

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.5986. Min Act: -0.0000

Data Index: 545726 (C4 (Web Text))

Max Activating Token Index: 386

Text #14

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.4946. Min Act: -0.0000

Data Index: 389407 (C4 (Web Text))

Max Activating Token Index: 459

Text #15

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.5997. Min Act: -0.0000

Data Index: 730398 (C4 (Web Text))

Max Activating Token Index: 258

Text #16

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.5463. Min Act: -0.0000

Data Index: 16136 (C4 (Web Text))

Max Activating Token Index: 3

Text #17

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.5739. Min Act: -0.0000

Data Index: 1127563 (C4 (Web Text))

Max Activating Token Index: 320

Text #18

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.5909. Min Act: -0.0000

Data Index: 1431115 (Python Code)

Max Activating Token Index: 437

Text #19

Max Range: 2.0464. Min Range: -2.0464

Max Act: 1.4664. Min Act: -0.0000

Data Index: 167859 (C4 (Web Text))

Max Activating Token Index: 156