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

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

Neuron 1365 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.2846. Min Range: -2.2846

Max Act: 2.0900. Min Act: -0.0000

Data Index: 56344 (C4 (Web Text))

Max Activating Token Index: 12

Text #1

Max Range: 2.2846. Min Range: -2.2846

Max Act: 2.2846. Min Act: -0.0000

Data Index: 983488 (C4 (Web Text))

Max Activating Token Index: 729

Text #2

Max Range: 2.2846. Min Range: -2.2846

Max Act: 2.0795. Min Act: -0.0000

Data Index: 96848 (C4 (Web Text))

Max Activating Token Index: 121

Text #3

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.8920. Min Act: -0.0000

Data Index: 237634 (C4 (Web Text))

Max Activating Token Index: 265

Text #4

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.9891. Min Act: -0.0000

Data Index: 1434561 (Python Code)

Max Activating Token Index: 798

Text #5

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.9335. Min Act: -0.0000

Data Index: 972150 (C4 (Web Text))

Max Activating Token Index: 766

Text #6

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.9015. Min Act: -0.0000

Data Index: 715401 (C4 (Web Text))

Max Activating Token Index: 64

Text #7

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.9860. Min Act: -0.0000

Data Index: 1343837 (C4 (Web Text))

Max Activating Token Index: 282

Text #8

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.8573. Min Act: -0.0000

Data Index: 550896 (C4 (Web Text))

Max Activating Token Index: 502

Text #9

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.8321. Min Act: -0.0000

Data Index: 880065 (C4 (Web Text))

Max Activating Token Index: 129

Text #10

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.8081. Min Act: -0.0000

Data Index: 1412477 (Python Code)

Max Activating Token Index: 329

Text #11

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.8916. Min Act: -0.0000

Data Index: 1228743 (C4 (Web Text))

Max Activating Token Index: 999

Text #12

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.6861. Min Act: -0.0000

Data Index: 598427 (C4 (Web Text))

Max Activating Token Index: 44

Text #13

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.8817. Min Act: -0.0000

Data Index: 250274 (C4 (Web Text))

Max Activating Token Index: 1001

Text #14

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.6685. Min Act: -0.0000

Data Index: 1150105 (C4 (Web Text))

Max Activating Token Index: 205

Text #15

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.6858. Min Act: -0.0000

Data Index: 1054734 (C4 (Web Text))

Max Activating Token Index: 338

Text #16

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.4756. Min Act: -0.0000

Data Index: 353263 (C4 (Web Text))

Max Activating Token Index: 423

Text #17

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.7275. Min Act: -0.0000

Data Index: 1519945 (Python Code)

Max Activating Token Index: 467

Text #18

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.7162. Min Act: -0.0000

Data Index: 1241161 (C4 (Web Text))

Max Activating Token Index: 140

Text #19

Max Range: 2.2846. Min Range: -2.2846

Max Act: 1.6999. Min Act: -0.0000

Data Index: 268636 (C4 (Web Text))

Max Activating Token Index: 345