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

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

Neuron 4979 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.2963. Min Range: -6.2963

Max Act: 6.2963. Min Act: 0.0000

Data Index: 1387360 (Python Code)

Max Activating Token Index: 250

Text #1

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.8530. Min Act: -0.0000

Data Index: 1514448 (Python Code)

Max Activating Token Index: 223

Text #2

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.6612. Min Act: -0.0000

Data Index: 1551730 (Python Code)

Max Activating Token Index: 278

Text #3

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.6161. Min Act: 0.0000

Data Index: 1382961 (Python Code)

Max Activating Token Index: 194

Text #4

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.0055. Min Act: -0.0000

Data Index: 1392011 (Python Code)

Max Activating Token Index: 599

Text #5

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.5457. Min Act: -0.0000

Data Index: 1457400 (Python Code)

Max Activating Token Index: 282

Text #6

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.3242. Min Act: 0.0000

Data Index: 1451418 (Python Code)

Max Activating Token Index: 251

Text #7

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.0351. Min Act: 0.0000

Data Index: 1557039 (Python Code)

Max Activating Token Index: 974

Text #8

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.2273. Min Act: -0.0000

Data Index: 1424733 (Python Code)

Max Activating Token Index: 648

Text #9

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.0117. Min Act: 0.0000

Data Index: 1538435 (Python Code)

Max Activating Token Index: 262

Text #10

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.2625. Min Act: 0.0000

Data Index: 1407305 (Python Code)

Max Activating Token Index: 188

Text #11

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.0257. Min Act: -0.0000

Data Index: 1465203 (Python Code)

Max Activating Token Index: 341

Text #12

Max Range: 6.2963. Min Range: -6.2963

Max Act: 4.9710. Min Act: 0.0000

Data Index: 1499175 (Python Code)

Max Activating Token Index: 80

Text #13

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.1475. Min Act: 0.0000

Data Index: 1380468 (Python Code)

Max Activating Token Index: 786

Text #14

Max Range: 6.2963. Min Range: -6.2963

Max Act: 5.1084. Min Act: 0.0000

Data Index: 1391938 (Python Code)

Max Activating Token Index: 968

Text #15

Max Range: 6.2963. Min Range: -6.2963

Max Act: 4.8887. Min Act: -0.0000

Data Index: 1560455 (Python Code)

Max Activating Token Index: 192

Text #16

Max Range: 6.2963. Min Range: -6.2963

Max Act: 4.7242. Min Act: -0.0000

Data Index: 1441651 (Python Code)

Max Activating Token Index: 132

Text #17

Max Range: 6.2963. Min Range: -6.2963

Max Act: 4.8449. Min Act: 0.0000

Data Index: 1417686 (Python Code)

Max Activating Token Index: 534

Text #18

Max Range: 6.2963. Min Range: -6.2963

Max Act: 4.6797. Min Act: -0.0000

Data Index: 1528037 (Python Code)

Max Activating Token Index: 823

Text #19

Max Range: 6.2963. Min Range: -6.2963

Max Act: 4.8726. Min Act: -0.0000

Data Index: 1413354 (Python Code)

Max Activating Token Index: 107