Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3563 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.9053. Min Act: -0.0002

Data Index: 515779 (C4 (Web Text))

Max Activating Token Index: 33

Text #1

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.8295. Min Act: -0.0002

Data Index: 1162858 (C4 (Web Text))

Max Activating Token Index: 253

Text #2

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.8571. Min Act: -0.0002

Data Index: 1574885 (Python Code)

Max Activating Token Index: 739

Text #3

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.8315. Min Act: -0.0002

Data Index: 395909 (C4 (Web Text))

Max Activating Token Index: 456

Text #4

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.8196. Min Act: -0.0002

Data Index: 1028402 (C4 (Web Text))

Max Activating Token Index: 474

Text #5

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7614. Min Act: -0.0002

Data Index: 1338339 (C4 (Web Text))

Max Activating Token Index: 466

Text #6

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.8081. Min Act: -0.0002

Data Index: 404717 (C4 (Web Text))

Max Activating Token Index: 419

Text #7

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7314. Min Act: -0.0002

Data Index: 948832 (C4 (Web Text))

Max Activating Token Index: 544

Text #8

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7955. Min Act: -0.0002

Data Index: 1397030 (Python Code)

Max Activating Token Index: 370

Text #9

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7944. Min Act: -0.0002

Data Index: 1116818 (C4 (Web Text))

Max Activating Token Index: 723

Text #10

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7390. Min Act: -0.0002

Data Index: 238618 (C4 (Web Text))

Max Activating Token Index: 691

Text #11

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7452. Min Act: -0.0002

Data Index: 1537067 (Python Code)

Max Activating Token Index: 215

Text #12

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7018. Min Act: -0.0002

Data Index: 1149862 (C4 (Web Text))

Max Activating Token Index: 204

Text #13

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7215. Min Act: -0.0002

Data Index: 633029 (C4 (Web Text))

Max Activating Token Index: 367

Text #14

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.6722. Min Act: -0.0002

Data Index: 854922 (C4 (Web Text))

Max Activating Token Index: 199

Text #15

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7566. Min Act: -0.0002

Data Index: 80008 (C4 (Web Text))

Max Activating Token Index: 526

Text #16

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7594. Min Act: -0.0002

Data Index: 730048 (C4 (Web Text))

Max Activating Token Index: 272

Text #17

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7293. Min Act: -0.0002

Data Index: 1226819 (C4 (Web Text))

Max Activating Token Index: 180

Text #18

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7481. Min Act: -0.0002

Data Index: 462076 (C4 (Web Text))

Max Activating Token Index: 3

Text #19

Max Range: 0.9053. Min Range: -0.9053

Max Act: 0.7263. Min Act: -0.0002

Data Index: 1362653 (Python Code)

Max Activating Token Index: 534