Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4017 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.0040. Min Range: -4.0040

Max Act: 4.0040. Min Act: -0.0000

Data Index: 1125254 (C4 (Web Text))

Max Activating Token Index: 275

Text #1

Max Range: 4.0040. Min Range: -4.0040

Max Act: 3.2936. Min Act: -0.0000

Data Index: 1455813 (Python Code)

Max Activating Token Index: 191

Text #2

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.8064. Min Act: -0.0000

Data Index: 1088670 (C4 (Web Text))

Max Activating Token Index: 588

Text #3

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.4645. Min Act: -0.0000

Data Index: 1400744 (Python Code)

Max Activating Token Index: 602

Text #4

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.4742. Min Act: -0.0000

Data Index: 1645689 (Python Code)

Max Activating Token Index: 517

Text #5

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.4196. Min Act: -0.0000

Data Index: 1430316 (Python Code)

Max Activating Token Index: 590

Text #6

Max Range: 4.0040. Min Range: -4.0040

Max Act: 1.9585. Min Act: -0.0000

Data Index: 387391 (C4 (Web Text))

Max Activating Token Index: 1003

Text #7

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.1544. Min Act: -0.0000

Data Index: 420077 (C4 (Web Text))

Max Activating Token Index: 707

Text #8

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.2706. Min Act: -0.0000

Data Index: 253030 (C4 (Web Text))

Max Activating Token Index: 887

Text #9

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.1551. Min Act: -0.0000

Data Index: 1070641 (C4 (Web Text))

Max Activating Token Index: 1010

Text #10

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.1729. Min Act: -0.0000

Data Index: 1433671 (Python Code)

Max Activating Token Index: 776

Text #11

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.1193. Min Act: -0.0000

Data Index: 638713 (C4 (Web Text))

Max Activating Token Index: 383

Text #12

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.1210. Min Act: -0.0000

Data Index: 1543163 (Python Code)

Max Activating Token Index: 978

Text #13

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.1101. Min Act: -0.0000

Data Index: 629180 (C4 (Web Text))

Max Activating Token Index: 1006

Text #14

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.0416. Min Act: -0.0000

Data Index: 1554385 (Python Code)

Max Activating Token Index: 146

Text #15

Max Range: 4.0040. Min Range: -4.0040

Max Act: 2.0106. Min Act: -0.0000

Data Index: 74768 (C4 (Web Text))

Max Activating Token Index: 555

Text #16

Max Range: 4.0040. Min Range: -4.0040

Max Act: 1.9750. Min Act: -0.0000

Data Index: 745076 (C4 (Web Text))

Max Activating Token Index: 730

Text #17

Max Range: 4.0040. Min Range: -4.0040

Max Act: 1.9527. Min Act: -0.0000

Data Index: 466150 (C4 (Web Text))

Max Activating Token Index: 363

Text #18

Max Range: 4.0040. Min Range: -4.0040

Max Act: 1.8470. Min Act: -0.0000

Data Index: 1059109 (C4 (Web Text))

Max Activating Token Index: 565

Text #19

Max Range: 4.0040. Min Range: -4.0040

Max Act: 1.9130. Min Act: -0.0000

Data Index: 729514 (C4 (Web Text))

Max Activating Token Index: 367