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

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

Neuron 2138 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: 0.4696. Min Range: -0.4696

Max Act: 0.4696. Min Act: -0.0001

Data Index: 1529873 (Python Code)

Max Activating Token Index: 63

Text #1

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.3149. Min Act: -0.0001

Data Index: 1479422 (Python Code)

Max Activating Token Index: 102

Text #2

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.3208. Min Act: -0.0001

Data Index: 761699 (C4 (Web Text))

Max Activating Token Index: 575

Text #3

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2864. Min Act: -0.0001

Data Index: 675462 (C4 (Web Text))

Max Activating Token Index: 410

Text #4

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2815. Min Act: -0.0001

Data Index: 437638 (C4 (Web Text))

Max Activating Token Index: 170

Text #5

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2719. Min Act: -0.0001

Data Index: 44109 (C4 (Web Text))

Max Activating Token Index: 214

Text #6

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2529. Min Act: -0.0001

Data Index: 648960 (C4 (Web Text))

Max Activating Token Index: 757

Text #7

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2841. Min Act: -0.0001

Data Index: 1161088 (C4 (Web Text))

Max Activating Token Index: 802

Text #8

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2638. Min Act: -0.0001

Data Index: 1109855 (C4 (Web Text))

Max Activating Token Index: 196

Text #9

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2826. Min Act: -0.0001

Data Index: 1145125 (C4 (Web Text))

Max Activating Token Index: 955

Text #10

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2539. Min Act: -0.0001

Data Index: 93614 (C4 (Web Text))

Max Activating Token Index: 995

Text #11

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2750. Min Act: -0.0001

Data Index: 1238800 (C4 (Web Text))

Max Activating Token Index: 345

Text #12

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2737. Min Act: -0.0001

Data Index: 1353479 (C4 (Web Text))

Max Activating Token Index: 570

Text #13

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2443. Min Act: -0.0001

Data Index: 1599173 (Python Code)

Max Activating Token Index: 274

Text #14

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2712. Min Act: -0.0001

Data Index: 1498265 (Python Code)

Max Activating Token Index: 472

Text #15

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2393. Min Act: -0.0001

Data Index: 373309 (C4 (Web Text))

Max Activating Token Index: 251

Text #16

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2395. Min Act: -0.0001

Data Index: 211226 (C4 (Web Text))

Max Activating Token Index: 89

Text #17

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2454. Min Act: -0.0001

Data Index: 613459 (C4 (Web Text))

Max Activating Token Index: 258

Text #18

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2365. Min Act: -0.0001

Data Index: 881718 (C4 (Web Text))

Max Activating Token Index: 97

Text #19

Max Range: 0.4696. Min Range: -0.4696

Max Act: 0.2453. Min Act: -0.0001

Data Index: 1146889 (C4 (Web Text))

Max Activating Token Index: 336