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

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

Neuron 2991 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4857. Min Range: -3.4857

Max Act: 2.8668. Min Act: -0.0000

Data Index: 1619873 (Python Code)

Max Activating Token Index: 706

Text #1

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.2081. Min Act: -0.0000

Data Index: 1616846 (Python Code)

Max Activating Token Index: 267

Text #2

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.0019. Min Act: -0.0000

Data Index: 1378607 (Python Code)

Max Activating Token Index: 750

Text #3

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.1522. Min Act: -0.0000

Data Index: 458466 (C4 (Web Text))

Max Activating Token Index: 381

Text #4

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.2904. Min Act: -0.0000

Data Index: 1522881 (Python Code)

Max Activating Token Index: 435

Text #5

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.3055. Min Act: -0.0000

Data Index: 127338 (C4 (Web Text))

Max Activating Token Index: 1004

Text #6

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.2213. Min Act: -0.0000

Data Index: 1440683 (Python Code)

Max Activating Token Index: 500

Text #7

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.3834. Min Act: -0.0000

Data Index: 1058672 (C4 (Web Text))

Max Activating Token Index: 245

Text #8

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.1485. Min Act: -0.0000

Data Index: 1541844 (Python Code)

Max Activating Token Index: 341

Text #9

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.4857. Min Act: -0.0000

Data Index: 1203481 (C4 (Web Text))

Max Activating Token Index: 906

Text #10

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.1705. Min Act: -0.0000

Data Index: 954356 (C4 (Web Text))

Max Activating Token Index: 777

Text #11

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.3094. Min Act: -0.0000

Data Index: 764304 (C4 (Web Text))

Max Activating Token Index: 970

Text #12

Max Range: 3.4857. Min Range: -3.4857

Max Act: 2.8253. Min Act: -0.0000

Data Index: 40521 (C4 (Web Text))

Max Activating Token Index: 730

Text #13

Max Range: 3.4857. Min Range: -3.4857

Max Act: 2.7366. Min Act: -0.0000

Data Index: 1545683 (Python Code)

Max Activating Token Index: 905

Text #14

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.0612. Min Act: -0.0000

Data Index: 1360031 (Python Code)

Max Activating Token Index: 631

Text #15

Max Range: 3.4857. Min Range: -3.4857

Max Act: 2.8988. Min Act: -0.0000

Data Index: 524807 (C4 (Web Text))

Max Activating Token Index: 780

Text #16

Max Range: 3.4857. Min Range: -3.4857

Max Act: 2.9997. Min Act: -0.0000

Data Index: 1632307 (Python Code)

Max Activating Token Index: 701

Text #17

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.0160. Min Act: -0.0000

Data Index: 1505043 (Python Code)

Max Activating Token Index: 162

Text #18

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.1632. Min Act: -0.0000

Data Index: 1181562 (C4 (Web Text))

Max Activating Token Index: 365

Text #19

Max Range: 3.4857. Min Range: -3.4857

Max Act: 3.1487. Min Act: -0.0000

Data Index: 372229 (C4 (Web Text))

Max Activating Token Index: 347