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

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

Neuron 5305 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.5438. Min Act: -0.0000

Data Index: 1508941 (Python Code)

Max Activating Token Index: 347

Text #1

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.4884. Min Act: -0.0000

Data Index: 1416225 (Python Code)

Max Activating Token Index: 511

Text #2

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.4097. Min Act: -0.0000

Data Index: 828262 (C4 (Web Text))

Max Activating Token Index: 137

Text #3

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3336. Min Act: -0.0000

Data Index: 1404606 (Python Code)

Max Activating Token Index: 343

Text #4

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3859. Min Act: -0.0001

Data Index: 1162546 (C4 (Web Text))

Max Activating Token Index: 463

Text #5

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3467. Min Act: -0.0001

Data Index: 949511 (C4 (Web Text))

Max Activating Token Index: 765

Text #6

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3585. Min Act: -0.0000

Data Index: 1440939 (Python Code)

Max Activating Token Index: 393

Text #7

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3435. Min Act: -0.0000

Data Index: 1072376 (C4 (Web Text))

Max Activating Token Index: 77

Text #8

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3450. Min Act: -0.0001

Data Index: 519720 (C4 (Web Text))

Max Activating Token Index: 122

Text #9

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3529. Min Act: -0.0001

Data Index: 72626 (C4 (Web Text))

Max Activating Token Index: 713

Text #10

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3282. Min Act: -0.0000

Data Index: 1395516 (Python Code)

Max Activating Token Index: 361

Text #11

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3142. Min Act: -0.0000

Data Index: 531630 (C4 (Web Text))

Max Activating Token Index: 801

Text #12

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3089. Min Act: -0.0000

Data Index: 230784 (C4 (Web Text))

Max Activating Token Index: 926

Text #13

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.2866. Min Act: -0.0000

Data Index: 1361372 (Python Code)

Max Activating Token Index: 756

Text #14

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.2974. Min Act: -0.0001

Data Index: 771587 (C4 (Web Text))

Max Activating Token Index: 345

Text #15

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.2973. Min Act: -0.0000

Data Index: 1485803 (Python Code)

Max Activating Token Index: 550

Text #16

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3280. Min Act: -0.0000

Data Index: 1040374 (C4 (Web Text))

Max Activating Token Index: 520

Text #17

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.3209. Min Act: -0.0001

Data Index: 733024 (C4 (Web Text))

Max Activating Token Index: 264

Text #18

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.2942. Min Act: -0.0001

Data Index: 1231636 (C4 (Web Text))

Max Activating Token Index: 368

Text #19

Max Range: 0.5438. Min Range: -0.5438

Max Act: 0.2973. Min Act: -0.0000

Data Index: 1505753 (Python Code)

Max Activating Token Index: 890