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

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

Neuron 154 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: 4.6862. Min Range: -4.6862

Max Act: 4.6862. Min Act: -0.0000

Data Index: 1200278 (C4 (Web Text))

Max Activating Token Index: 676

Text #1

Max Range: 4.6862. Min Range: -4.6862

Max Act: 3.7420. Min Act: -0.0000

Data Index: 1297261 (C4 (Web Text))

Max Activating Token Index: 630

Text #2

Max Range: 4.6862. Min Range: -4.6862

Max Act: 3.1858. Min Act: -0.0000

Data Index: 543887 (C4 (Web Text))

Max Activating Token Index: 70

Text #3

Max Range: 4.6862. Min Range: -4.6862

Max Act: 3.3552. Min Act: -0.0000

Data Index: 197648 (C4 (Web Text))

Max Activating Token Index: 253

Text #4

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.5916. Min Act: -0.0000

Data Index: 900756 (C4 (Web Text))

Max Activating Token Index: 540

Text #5

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.8161. Min Act: -0.0000

Data Index: 544902 (C4 (Web Text))

Max Activating Token Index: 137

Text #6

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.6691. Min Act: -0.0000

Data Index: 276762 (C4 (Web Text))

Max Activating Token Index: 138

Text #7

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.7676. Min Act: -0.0000

Data Index: 1289164 (C4 (Web Text))

Max Activating Token Index: 1006

Text #8

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.7913. Min Act: -0.0000

Data Index: 264182 (C4 (Web Text))

Max Activating Token Index: 248

Text #9

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.7510. Min Act: -0.0000

Data Index: 606167 (C4 (Web Text))

Max Activating Token Index: 522

Text #10

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.4268. Min Act: -0.0000

Data Index: 149082 (C4 (Web Text))

Max Activating Token Index: 617

Text #11

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.6380. Min Act: -0.0000

Data Index: 960939 (C4 (Web Text))

Max Activating Token Index: 805

Text #12

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.6720. Min Act: -0.0000

Data Index: 485632 (C4 (Web Text))

Max Activating Token Index: 272

Text #13

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.6639. Min Act: -0.0000

Data Index: 529731 (C4 (Web Text))

Max Activating Token Index: 522

Text #14

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.5407. Min Act: -0.0000

Data Index: 32435 (C4 (Web Text))

Max Activating Token Index: 645

Text #15

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.4866. Min Act: -0.0000

Data Index: 231439 (C4 (Web Text))

Max Activating Token Index: 155

Text #16

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.3229. Min Act: -0.0000

Data Index: 793597 (C4 (Web Text))

Max Activating Token Index: 185

Text #17

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.4507. Min Act: -0.0000

Data Index: 1257171 (C4 (Web Text))

Max Activating Token Index: 645

Text #18

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.2926. Min Act: -0.0000

Data Index: 987629 (C4 (Web Text))

Max Activating Token Index: 174

Text #19

Max Range: 4.6862. Min Range: -4.6862

Max Act: 2.3882. Min Act: -0.0000

Data Index: 34504 (C4 (Web Text))

Max Activating Token Index: 904