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

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

Neuron 761 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0525. Min Act: -0.0001

Data Index: 642185 (C4 (Web Text))

Max Activating Token Index: 527

Text #1

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0455. Min Act: -0.0001

Data Index: 1347917 (C4 (Web Text))

Max Activating Token Index: 167

Text #2

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0461. Min Act: -0.0001

Data Index: 187808 (C4 (Web Text))

Max Activating Token Index: 209

Text #3

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0394. Min Act: -0.0001

Data Index: 103632 (C4 (Web Text))

Max Activating Token Index: 951

Text #4

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0401. Min Act: -0.0001

Data Index: 62122 (C4 (Web Text))

Max Activating Token Index: 342

Text #5

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0338. Min Act: -0.0001

Data Index: 867464 (C4 (Web Text))

Max Activating Token Index: 811

Text #6

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0348. Min Act: -0.0001

Data Index: 462721 (C4 (Web Text))

Max Activating Token Index: 383

Text #7

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0376. Min Act: -0.0001

Data Index: 1267600 (C4 (Web Text))

Max Activating Token Index: 436

Text #8

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0334. Min Act: -0.0001

Data Index: 1572927 (Python Code)

Max Activating Token Index: 115

Text #9

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0361. Min Act: -0.0001

Data Index: 771962 (C4 (Web Text))

Max Activating Token Index: 617

Text #10

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0327. Min Act: -0.0001

Data Index: 166111 (C4 (Web Text))

Max Activating Token Index: 701

Text #11

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0333. Min Act: -0.0001

Data Index: 131216 (C4 (Web Text))

Max Activating Token Index: 356

Text #12

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0307. Min Act: -0.0001

Data Index: 483531 (C4 (Web Text))

Max Activating Token Index: 843

Text #13

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0329. Min Act: -0.0001

Data Index: 1438581 (Python Code)

Max Activating Token Index: 335

Text #14

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0314. Min Act: -0.0001

Data Index: 38413 (C4 (Web Text))

Max Activating Token Index: 542

Text #15

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0337. Min Act: -0.0001

Data Index: 137539 (C4 (Web Text))

Max Activating Token Index: 225

Text #16

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0309. Min Act: -0.0001

Data Index: 1317603 (C4 (Web Text))

Max Activating Token Index: 287

Text #17

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0322. Min Act: -0.0001

Data Index: 342802 (C4 (Web Text))

Max Activating Token Index: 732

Text #18

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0346. Min Act: -0.0001

Data Index: 1293670 (C4 (Web Text))

Max Activating Token Index: 936

Text #19

Max Range: 0.0525. Min Range: -0.0525

Max Act: 0.0333. Min Act: -0.0001

Data Index: 32332 (C4 (Web Text))

Max Activating Token Index: 889