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

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

Neuron 4533 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: 6.8523. Min Range: -6.8523

Max Act: 6.8523. Min Act: -0.0000

Data Index: 1238525 (C4 (Web Text))

Max Activating Token Index: 319

Text #1

Max Range: 6.8523. Min Range: -6.8523

Max Act: 6.5942. Min Act: -0.0000

Data Index: 51250 (C4 (Web Text))

Max Activating Token Index: 774

Text #2

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.7200. Min Act: -0.0000

Data Index: 443500 (C4 (Web Text))

Max Activating Token Index: 764

Text #3

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.6261. Min Act: -0.0000

Data Index: 524095 (C4 (Web Text))

Max Activating Token Index: 149

Text #4

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.7942. Min Act: -0.0000

Data Index: 1320223 (C4 (Web Text))

Max Activating Token Index: 883

Text #5

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.4644. Min Act: -0.0000

Data Index: 1366298 (Python Code)

Max Activating Token Index: 880

Text #6

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.0686. Min Act: -0.0000

Data Index: 502531 (C4 (Web Text))

Max Activating Token Index: 590

Text #7

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.1541. Min Act: -0.0000

Data Index: 1405944 (Python Code)

Max Activating Token Index: 663

Text #8

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.2495. Min Act: -0.0000

Data Index: 969332 (C4 (Web Text))

Max Activating Token Index: 1018

Text #9

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.2013. Min Act: -0.0000

Data Index: 1576589 (Python Code)

Max Activating Token Index: 892

Text #10

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.1729. Min Act: -0.0000

Data Index: 1409478 (Python Code)

Max Activating Token Index: 231

Text #11

Max Range: 6.8523. Min Range: -6.8523

Max Act: 4.9296. Min Act: -0.0000

Data Index: 983778 (C4 (Web Text))

Max Activating Token Index: 993

Text #12

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.1416. Min Act: -0.0000

Data Index: 399360 (C4 (Web Text))

Max Activating Token Index: 899

Text #13

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.1330. Min Act: -0.0000

Data Index: 1458027 (Python Code)

Max Activating Token Index: 634

Text #14

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.0680. Min Act: -0.0000

Data Index: 32982 (C4 (Web Text))

Max Activating Token Index: 966

Text #15

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.1430. Min Act: -0.0000

Data Index: 1141349 (C4 (Web Text))

Max Activating Token Index: 1011

Text #16

Max Range: 6.8523. Min Range: -6.8523

Max Act: 4.9437. Min Act: -0.0000

Data Index: 1370569 (Python Code)

Max Activating Token Index: 937

Text #17

Max Range: 6.8523. Min Range: -6.8523

Max Act: 4.9459. Min Act: -0.0000

Data Index: 1467133 (Python Code)

Max Activating Token Index: 678

Text #18

Max Range: 6.8523. Min Range: -6.8523

Max Act: 4.9338. Min Act: -0.0000

Data Index: 1642207 (Python Code)

Max Activating Token Index: 447

Text #19

Max Range: 6.8523. Min Range: -6.8523

Max Act: 5.0920. Min Act: -0.0000

Data Index: 936850 (C4 (Web Text))

Max Activating Token Index: 14