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

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

Neuron 4620 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.6757. Min Act: -0.0000

Data Index: 651212 (C4 (Web Text))

Max Activating Token Index: 528

Text #1

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.6349. Min Act: -0.0000

Data Index: 217482 (C4 (Web Text))

Max Activating Token Index: 309

Text #2

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.5070. Min Act: -0.0000

Data Index: 695239 (C4 (Web Text))

Max Activating Token Index: 433

Text #3

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.4524. Min Act: -0.0000

Data Index: 612776 (C4 (Web Text))

Max Activating Token Index: 951

Text #4

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.4512. Min Act: 0.0001

Data Index: 1466834 (Python Code)

Max Activating Token Index: 556

Text #5

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.4009. Min Act: -0.0000

Data Index: 225067 (C4 (Web Text))

Max Activating Token Index: 453

Text #6

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.4217. Min Act: -0.0000

Data Index: 67456 (C4 (Web Text))

Max Activating Token Index: 589

Text #7

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.4163. Min Act: -0.0000

Data Index: 55831 (C4 (Web Text))

Max Activating Token Index: 97

Text #8

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3972. Min Act: -0.0000

Data Index: 917011 (C4 (Web Text))

Max Activating Token Index: 537

Text #9

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3884. Min Act: -0.0000

Data Index: 854206 (C4 (Web Text))

Max Activating Token Index: 481

Text #10

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3664. Min Act: 0.0000

Data Index: 1491741 (Python Code)

Max Activating Token Index: 945

Text #11

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3878. Min Act: -0.0000

Data Index: 237011 (C4 (Web Text))

Max Activating Token Index: 536

Text #12

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3388. Min Act: 0.0000

Data Index: 1646115 (Python Code)

Max Activating Token Index: 255

Text #13

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3510. Min Act: -0.0000

Data Index: 62105 (C4 (Web Text))

Max Activating Token Index: 859

Text #14

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3665. Min Act: 0.0000

Data Index: 1412594 (Python Code)

Max Activating Token Index: 259

Text #15

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3601. Min Act: -0.0000

Data Index: 1221530 (C4 (Web Text))

Max Activating Token Index: 194

Text #16

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.4217. Min Act: -0.0000

Data Index: 754192 (C4 (Web Text))

Max Activating Token Index: 907

Text #17

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3289. Min Act: -0.0000

Data Index: 1116443 (C4 (Web Text))

Max Activating Token Index: 752

Text #18

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3538. Min Act: -0.0000

Data Index: 648154 (C4 (Web Text))

Max Activating Token Index: 915

Text #19

Max Range: 0.6757. Min Range: -0.6757

Max Act: 0.3489. Min Act: -0.0000

Data Index: 104640 (C4 (Web Text))

Max Activating Token Index: 375