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

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

Neuron 2854 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0695. Min Act: -0.0001

Data Index: 650488 (C4 (Web Text))

Max Activating Token Index: 937

Text #1

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0652. Min Act: -0.0001

Data Index: 727979 (C4 (Web Text))

Max Activating Token Index: 105

Text #2

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0591. Min Act: -0.0001

Data Index: 1075825 (C4 (Web Text))

Max Activating Token Index: 765

Text #3

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0582. Min Act: -0.0001

Data Index: 121112 (C4 (Web Text))

Max Activating Token Index: 926

Text #4

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0540. Min Act: -0.0001

Data Index: 530499 (C4 (Web Text))

Max Activating Token Index: 84

Text #5

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0545. Min Act: -0.0001

Data Index: 405152 (C4 (Web Text))

Max Activating Token Index: 284

Text #6

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0514. Min Act: -0.0001

Data Index: 291842 (C4 (Web Text))

Max Activating Token Index: 516

Text #7

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0463. Min Act: -0.0001

Data Index: 904007 (C4 (Web Text))

Max Activating Token Index: 453

Text #8

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0481. Min Act: -0.0001

Data Index: 711765 (C4 (Web Text))

Max Activating Token Index: 50

Text #9

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0487. Min Act: -0.0001

Data Index: 838614 (C4 (Web Text))

Max Activating Token Index: 919

Text #10

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0442. Min Act: -0.0001

Data Index: 1052745 (C4 (Web Text))

Max Activating Token Index: 780

Text #11

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0456. Min Act: -0.0001

Data Index: 1320988 (C4 (Web Text))

Max Activating Token Index: 773

Text #12

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0466. Min Act: -0.0001

Data Index: 777102 (C4 (Web Text))

Max Activating Token Index: 658

Text #13

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0463. Min Act: -0.0001

Data Index: 635205 (C4 (Web Text))

Max Activating Token Index: 547

Text #14

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0461. Min Act: -0.0001

Data Index: 571540 (C4 (Web Text))

Max Activating Token Index: 441

Text #15

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0454. Min Act: -0.0001

Data Index: 1642186 (Python Code)

Max Activating Token Index: 466

Text #16

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0483. Min Act: -0.0001

Data Index: 705835 (C4 (Web Text))

Max Activating Token Index: 330

Text #17

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0472. Min Act: -0.0001

Data Index: 1231637 (C4 (Web Text))

Max Activating Token Index: 132

Text #18

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0459. Min Act: -0.0001

Data Index: 554820 (C4 (Web Text))

Max Activating Token Index: 622

Text #19

Max Range: 0.0695. Min Range: -0.0695

Max Act: 0.0457. Min Act: -0.0001

Data Index: 1013078 (C4 (Web Text))

Max Activating Token Index: 305