Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4604 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.6962. Min Range: -1.6962

Max Act: 1.6962. Min Act: -0.0000

Data Index: 807410 (C4 (Web Text))

Max Activating Token Index: 858

Text #1

Max Range: 1.6962. Min Range: -1.6962

Max Act: 1.1294. Min Act: -0.0000

Data Index: 282426 (C4 (Web Text))

Max Activating Token Index: 595

Text #2

Max Range: 1.6962. Min Range: -1.6962

Max Act: 1.1530. Min Act: -0.0000

Data Index: 1336769 (C4 (Web Text))

Max Activating Token Index: 934

Text #3

Max Range: 1.6962. Min Range: -1.6962

Max Act: 1.0661. Min Act: -0.0000

Data Index: 1286014 (C4 (Web Text))

Max Activating Token Index: 485

Text #4

Max Range: 1.6962. Min Range: -1.6962

Max Act: 1.1099. Min Act: -0.0000

Data Index: 161248 (C4 (Web Text))

Max Activating Token Index: 601

Text #5

Max Range: 1.6962. Min Range: -1.6962

Max Act: 1.0838. Min Act: -0.0000

Data Index: 811856 (C4 (Web Text))

Max Activating Token Index: 686

Text #6

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.9443. Min Act: -0.0000

Data Index: 949053 (C4 (Web Text))

Max Activating Token Index: 915

Text #7

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.9633. Min Act: -0.0000

Data Index: 920250 (C4 (Web Text))

Max Activating Token Index: 523

Text #8

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.9565. Min Act: -0.0000

Data Index: 814431 (C4 (Web Text))

Max Activating Token Index: 40

Text #9

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.9829. Min Act: -0.0000

Data Index: 666735 (C4 (Web Text))

Max Activating Token Index: 343

Text #10

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.9355. Min Act: -0.0000

Data Index: 935212 (C4 (Web Text))

Max Activating Token Index: 45

Text #11

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.9261. Min Act: -0.0000

Data Index: 233192 (C4 (Web Text))

Max Activating Token Index: 1017

Text #12

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.9154. Min Act: -0.0000

Data Index: 1171322 (C4 (Web Text))

Max Activating Token Index: 866

Text #13

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.9027. Min Act: -0.0000

Data Index: 65259 (C4 (Web Text))

Max Activating Token Index: 864

Text #14

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.8710. Min Act: -0.0000

Data Index: 443413 (C4 (Web Text))

Max Activating Token Index: 241

Text #15

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.8433. Min Act: -0.0000

Data Index: 631588 (C4 (Web Text))

Max Activating Token Index: 217

Text #16

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.8976. Min Act: -0.0000

Data Index: 941079 (C4 (Web Text))

Max Activating Token Index: 646

Text #17

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.8488. Min Act: -0.0000

Data Index: 8616 (C4 (Web Text))

Max Activating Token Index: 579

Text #18

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.8877. Min Act: -0.0000

Data Index: 271028 (C4 (Web Text))

Max Activating Token Index: 586

Text #19

Max Range: 1.6962. Min Range: -1.6962

Max Act: 0.8507. Min Act: -0.0000

Data Index: 566017 (C4 (Web Text))

Max Activating Token Index: 764