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

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

Neuron 5808 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0878. Min Act: -0.0001

Data Index: 267238 (C4 (Web Text))

Max Activating Token Index: 723

Text #1

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0934. Min Act: -0.0001

Data Index: 23248 (C4 (Web Text))

Max Activating Token Index: 792

Text #2

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0641. Min Act: -0.0001

Data Index: 880096 (C4 (Web Text))

Max Activating Token Index: 260

Text #3

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0836. Min Act: -0.0001

Data Index: 1239920 (C4 (Web Text))

Max Activating Token Index: 716

Text #4

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0684. Min Act: -0.0001

Data Index: 583714 (C4 (Web Text))

Max Activating Token Index: 859

Text #5

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0557. Min Act: -0.0001

Data Index: 1290779 (C4 (Web Text))

Max Activating Token Index: 730

Text #6

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0600. Min Act: -0.0001

Data Index: 780357 (C4 (Web Text))

Max Activating Token Index: 6

Text #7

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0657. Min Act: -0.0001

Data Index: 50450 (C4 (Web Text))

Max Activating Token Index: 299

Text #8

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0520. Min Act: -0.0001

Data Index: 1430894 (Python Code)

Max Activating Token Index: 665

Text #9

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0521. Min Act: -0.0001

Data Index: 1390743 (Python Code)

Max Activating Token Index: 512

Text #10

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0412. Min Act: -0.0001

Data Index: 735025 (C4 (Web Text))

Max Activating Token Index: 320

Text #11

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0504. Min Act: -0.0001

Data Index: 142362 (C4 (Web Text))

Max Activating Token Index: 255

Text #12

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0536. Min Act: -0.0001

Data Index: 4491 (C4 (Web Text))

Max Activating Token Index: 342

Text #13

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0491. Min Act: -0.0001

Data Index: 266885 (C4 (Web Text))

Max Activating Token Index: 226

Text #14

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0486. Min Act: -0.0001

Data Index: 1379565 (Python Code)

Max Activating Token Index: 708

Text #15

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0674. Min Act: -0.0001

Data Index: 871710 (C4 (Web Text))

Max Activating Token Index: 218

Text #16

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0487. Min Act: -0.0001

Data Index: 92802 (C4 (Web Text))

Max Activating Token Index: 571

Text #17

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0540. Min Act: -0.0001

Data Index: 454949 (C4 (Web Text))

Max Activating Token Index: 360

Text #18

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0519. Min Act: -0.0001

Data Index: 566163 (C4 (Web Text))

Max Activating Token Index: 579

Text #19

Max Range: 0.0934. Min Range: -0.0934

Max Act: 0.0459. Min Act: -0.0001

Data Index: 1504274 (Python Code)

Max Activating Token Index: 2