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

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

Neuron 3634 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: 0.1639. Min Range: -0.1639

Max Act: 0.1639. Min Act: -0.0001

Data Index: 999097 (C4 (Web Text))

Max Activating Token Index: 18

Text #1

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1577. Min Act: -0.0001

Data Index: 69454 (C4 (Web Text))

Max Activating Token Index: 442

Text #2

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1573. Min Act: -0.0001

Data Index: 477278 (C4 (Web Text))

Max Activating Token Index: 964

Text #3

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1521. Min Act: -0.0001

Data Index: 1331007 (C4 (Web Text))

Max Activating Token Index: 227

Text #4

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1533. Min Act: -0.0001

Data Index: 1245202 (C4 (Web Text))

Max Activating Token Index: 712

Text #5

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1525. Min Act: -0.0001

Data Index: 684608 (C4 (Web Text))

Max Activating Token Index: 900

Text #6

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1466. Min Act: -0.0001

Data Index: 1098944 (C4 (Web Text))

Max Activating Token Index: 136

Text #7

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1564. Min Act: -0.0001

Data Index: 266633 (C4 (Web Text))

Max Activating Token Index: 855

Text #8

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1477. Min Act: -0.0001

Data Index: 54193 (C4 (Web Text))

Max Activating Token Index: 323

Text #9

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1467. Min Act: -0.0001

Data Index: 312809 (C4 (Web Text))

Max Activating Token Index: 312

Text #10

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1403. Min Act: -0.0001

Data Index: 964054 (C4 (Web Text))

Max Activating Token Index: 58

Text #11

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1411. Min Act: -0.0001

Data Index: 1005046 (C4 (Web Text))

Max Activating Token Index: 251

Text #12

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1377. Min Act: -0.0001

Data Index: 852513 (C4 (Web Text))

Max Activating Token Index: 323

Text #13

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1339. Min Act: -0.0001

Data Index: 182040 (C4 (Web Text))

Max Activating Token Index: 746

Text #14

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1377. Min Act: -0.0001

Data Index: 405016 (C4 (Web Text))

Max Activating Token Index: 8

Text #15

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1371. Min Act: -0.0001

Data Index: 990022 (C4 (Web Text))

Max Activating Token Index: 204

Text #16

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1272. Min Act: -0.0001

Data Index: 256314 (C4 (Web Text))

Max Activating Token Index: 919

Text #17

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1297. Min Act: -0.0001

Data Index: 1083575 (C4 (Web Text))

Max Activating Token Index: 809

Text #18

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1288. Min Act: -0.0001

Data Index: 421841 (C4 (Web Text))

Max Activating Token Index: 1018

Text #19

Max Range: 0.1639. Min Range: -0.1639

Max Act: 0.1237. Min Act: -0.0001

Data Index: 1048573 (C4 (Web Text))

Max Activating Token Index: 924