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

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

Neuron 4639 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: 0.9525. Min Range: -0.9525

Max Act: 0.8994. Min Act: -0.0000

Data Index: 1112311 (C4 (Web Text))

Max Activating Token Index: 742

Text #1

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.9525. Min Act: -0.0000

Data Index: 878116 (C4 (Web Text))

Max Activating Token Index: 465

Text #2

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.9030. Min Act: -0.0000

Data Index: 1344658 (C4 (Web Text))

Max Activating Token Index: 590

Text #3

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.8329. Min Act: -0.0000

Data Index: 913241 (C4 (Web Text))

Max Activating Token Index: 28

Text #4

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.8405. Min Act: -0.0000

Data Index: 1163701 (C4 (Web Text))

Max Activating Token Index: 771

Text #5

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.8164. Min Act: -0.0000

Data Index: 1371882 (Python Code)

Max Activating Token Index: 200

Text #6

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7848. Min Act: -0.0000

Data Index: 1236258 (C4 (Web Text))

Max Activating Token Index: 742

Text #7

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7365. Min Act: -0.0000

Data Index: 1633689 (Python Code)

Max Activating Token Index: 733

Text #8

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7499. Min Act: -0.0000

Data Index: 179379 (C4 (Web Text))

Max Activating Token Index: 486

Text #9

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7262. Min Act: -0.0000

Data Index: 1009528 (C4 (Web Text))

Max Activating Token Index: 1022

Text #10

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7242. Min Act: -0.0000

Data Index: 1502696 (Python Code)

Max Activating Token Index: 770

Text #11

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7427. Min Act: -0.0000

Data Index: 373693 (C4 (Web Text))

Max Activating Token Index: 508

Text #12

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.6900. Min Act: -0.0000

Data Index: 283963 (C4 (Web Text))

Max Activating Token Index: 652

Text #13

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7297. Min Act: -0.0000

Data Index: 1579850 (Python Code)

Max Activating Token Index: 265

Text #14

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7932. Min Act: -0.0000

Data Index: 1454884 (Python Code)

Max Activating Token Index: 929

Text #15

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7218. Min Act: -0.0000

Data Index: 864713 (C4 (Web Text))

Max Activating Token Index: 45

Text #16

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7457. Min Act: -0.0000

Data Index: 97698 (C4 (Web Text))

Max Activating Token Index: 65

Text #17

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.6770. Min Act: -0.0000

Data Index: 1448344 (Python Code)

Max Activating Token Index: 44

Text #18

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.6491. Min Act: -0.0000

Data Index: 1045058 (C4 (Web Text))

Max Activating Token Index: 782

Text #19

Max Range: 0.9525. Min Range: -0.9525

Max Act: 0.7338. Min Act: -0.0000

Data Index: 117854 (C4 (Web Text))

Max Activating Token Index: 476