Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2396 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 1.0740. Min Range: -1.0740

Max Act: 1.0740. Min Act: -0.0001

Data Index: 1142582 (C4 (Web Text))

Max Activating Token Index: 1003

Text #1

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.9071. Min Act: -0.0001

Data Index: 103601 (C4 (Web Text))

Max Activating Token Index: 488

Text #2

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.7822. Min Act: -0.0001

Data Index: 318756 (C4 (Web Text))

Max Activating Token Index: 848

Text #3

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8559. Min Act: -0.0001

Data Index: 513679 (C4 (Web Text))

Max Activating Token Index: 196

Text #4

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8927. Min Act: -0.0001

Data Index: 987641 (C4 (Web Text))

Max Activating Token Index: 678

Text #5

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.9625. Min Act: -0.0001

Data Index: 257488 (C4 (Web Text))

Max Activating Token Index: 564

Text #6

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8673. Min Act: -0.0001

Data Index: 500215 (C4 (Web Text))

Max Activating Token Index: 558

Text #7

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8688. Min Act: -0.0001

Data Index: 1320392 (C4 (Web Text))

Max Activating Token Index: 123

Text #8

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.7785. Min Act: -0.0001

Data Index: 117639 (C4 (Web Text))

Max Activating Token Index: 539

Text #9

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8320. Min Act: -0.0001

Data Index: 663037 (C4 (Web Text))

Max Activating Token Index: 717

Text #10

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.7812. Min Act: -0.0001

Data Index: 244070 (C4 (Web Text))

Max Activating Token Index: 643

Text #11

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8488. Min Act: -0.0001

Data Index: 480098 (C4 (Web Text))

Max Activating Token Index: 195

Text #12

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8388. Min Act: -0.0001

Data Index: 394905 (C4 (Web Text))

Max Activating Token Index: 258

Text #13

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.7963. Min Act: -0.0001

Data Index: 1357472 (C4 (Web Text))

Max Activating Token Index: 728

Text #14

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8451. Min Act: -0.0001

Data Index: 1435147 (Python Code)

Max Activating Token Index: 542

Text #15

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8061. Min Act: -0.0001

Data Index: 1316450 (C4 (Web Text))

Max Activating Token Index: 250

Text #16

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.7893. Min Act: -0.0001

Data Index: 514421 (C4 (Web Text))

Max Activating Token Index: 440

Text #17

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.7547. Min Act: -0.0001

Data Index: 620792 (C4 (Web Text))

Max Activating Token Index: 283

Text #18

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8276. Min Act: -0.0001

Data Index: 136915 (C4 (Web Text))

Max Activating Token Index: 115

Text #19

Max Range: 1.0740. Min Range: -1.0740

Max Act: 0.8086. Min Act: -0.0001

Data Index: 276747 (C4 (Web Text))

Max Activating Token Index: 195