Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2251 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.5278. Min Act: -0.0001

Data Index: 567281 (C4 (Web Text))

Max Activating Token Index: 472

Text #1

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.5192. Min Act: -0.0001

Data Index: 121549 (C4 (Web Text))

Max Activating Token Index: 20

Text #2

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.5209. Min Act: -0.0001

Data Index: 55467 (C4 (Web Text))

Max Activating Token Index: 746

Text #3

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.5152. Min Act: -0.0001

Data Index: 852186 (C4 (Web Text))

Max Activating Token Index: 17

Text #4

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4726. Min Act: -0.0001

Data Index: 962145 (C4 (Web Text))

Max Activating Token Index: 97

Text #5

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4632. Min Act: -0.0001

Data Index: 1300059 (C4 (Web Text))

Max Activating Token Index: 419

Text #6

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4326. Min Act: -0.0001

Data Index: 725460 (C4 (Web Text))

Max Activating Token Index: 928

Text #7

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4387. Min Act: -0.0001

Data Index: 321888 (C4 (Web Text))

Max Activating Token Index: 735

Text #8

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4533. Min Act: -0.0001

Data Index: 116361 (C4 (Web Text))

Max Activating Token Index: 712

Text #9

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4382. Min Act: -0.0001

Data Index: 645287 (C4 (Web Text))

Max Activating Token Index: 741

Text #10

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4579. Min Act: -0.0001

Data Index: 210119 (C4 (Web Text))

Max Activating Token Index: 616

Text #11

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4347. Min Act: -0.0001

Data Index: 1138423 (C4 (Web Text))

Max Activating Token Index: 533

Text #12

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4469. Min Act: -0.0001

Data Index: 457802 (C4 (Web Text))

Max Activating Token Index: 729

Text #13

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4721. Min Act: -0.0001

Data Index: 495618 (C4 (Web Text))

Max Activating Token Index: 556

Text #14

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4276. Min Act: -0.0001

Data Index: 959997 (C4 (Web Text))

Max Activating Token Index: 71

Text #15

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4327. Min Act: -0.0001

Data Index: 253201 (C4 (Web Text))

Max Activating Token Index: 248

Text #16

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4333. Min Act: -0.0001

Data Index: 364205 (C4 (Web Text))

Max Activating Token Index: 826

Text #17

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4337. Min Act: -0.0001

Data Index: 833927 (C4 (Web Text))

Max Activating Token Index: 425

Text #18

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4178. Min Act: -0.0001

Data Index: 576772 (C4 (Web Text))

Max Activating Token Index: 16

Text #19

Max Range: 0.5278. Min Range: -0.5278

Max Act: 0.4431. Min Act: -0.0001

Data Index: 895727 (C4 (Web Text))

Max Activating Token Index: 811