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

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

Neuron 261 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.6458. Min Act: -0.0003

Data Index: 1010609 (C4 (Web Text))

Max Activating Token Index: 211

Text #1

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.6293. Min Act: -0.0003

Data Index: 901833 (C4 (Web Text))

Max Activating Token Index: 31

Text #2

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.6402. Min Act: -0.0003

Data Index: 379412 (C4 (Web Text))

Max Activating Token Index: 29

Text #3

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5775. Min Act: -0.0003

Data Index: 407942 (C4 (Web Text))

Max Activating Token Index: 68

Text #4

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.6336. Min Act: -0.0003

Data Index: 874820 (C4 (Web Text))

Max Activating Token Index: 243

Text #5

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.6155. Min Act: -0.0003

Data Index: 511028 (C4 (Web Text))

Max Activating Token Index: 218

Text #6

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.6140. Min Act: -0.0003

Data Index: 59278 (C4 (Web Text))

Max Activating Token Index: 64

Text #7

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5806. Min Act: -0.0003

Data Index: 1508789 (Python Code)

Max Activating Token Index: 401

Text #8

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5723. Min Act: -0.0003

Data Index: 1279094 (C4 (Web Text))

Max Activating Token Index: 68

Text #9

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5898. Min Act: -0.0003

Data Index: 1631537 (Python Code)

Max Activating Token Index: 399

Text #10

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5919. Min Act: -0.0003

Data Index: 1252363 (C4 (Web Text))

Max Activating Token Index: 67

Text #11

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.6275. Min Act: -0.0003

Data Index: 1139419 (C4 (Web Text))

Max Activating Token Index: 840

Text #12

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5818. Min Act: -0.0003

Data Index: 121015 (C4 (Web Text))

Max Activating Token Index: 385

Text #13

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5876. Min Act: -0.0003

Data Index: 53734 (C4 (Web Text))

Max Activating Token Index: 55

Text #14

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5455. Min Act: -0.0003

Data Index: 949984 (C4 (Web Text))

Max Activating Token Index: 995

Text #15

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5985. Min Act: -0.0003

Data Index: 1468859 (Python Code)

Max Activating Token Index: 950

Text #16

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5546. Min Act: -0.0003

Data Index: 195852 (C4 (Web Text))

Max Activating Token Index: 60

Text #17

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5554. Min Act: -0.0003

Data Index: 923004 (C4 (Web Text))

Max Activating Token Index: 762

Text #18

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5559. Min Act: -0.0003

Data Index: 39725 (C4 (Web Text))

Max Activating Token Index: 48

Text #19

Max Range: 0.6458. Min Range: -0.6458

Max Act: 0.5181. Min Act: -0.0003

Data Index: 982774 (C4 (Web Text))

Max Activating Token Index: 849