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

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

Neuron 5970 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.4053. Min Act: -0.0000

Data Index: 1291402 (C4 (Web Text))

Max Activating Token Index: 959

Text #1

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.2578. Min Act: -0.0000

Data Index: 1394142 (Python Code)

Max Activating Token Index: 52

Text #2

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.2406. Min Act: -0.0000

Data Index: 1430964 (Python Code)

Max Activating Token Index: 99

Text #3

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.2199. Min Act: -0.0000

Data Index: 1396224 (Python Code)

Max Activating Token Index: 36

Text #4

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.1332. Min Act: -0.0000

Data Index: 1212430 (C4 (Web Text))

Max Activating Token Index: 780

Text #5

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.1269. Min Act: -0.0000

Data Index: 1421528 (Python Code)

Max Activating Token Index: 49

Text #6

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.1582. Min Act: -0.0000

Data Index: 728434 (C4 (Web Text))

Max Activating Token Index: 879

Text #7

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.1352. Min Act: -0.0000

Data Index: 1566051 (Python Code)

Max Activating Token Index: 55

Text #8

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.1551. Min Act: -0.0000

Data Index: 700115 (C4 (Web Text))

Max Activating Token Index: 505

Text #9

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.1496. Min Act: -0.0000

Data Index: 1021496 (C4 (Web Text))

Max Activating Token Index: 941

Text #10

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.1103. Min Act: -0.0000

Data Index: 1650958 (Python Code)

Max Activating Token Index: 136

Text #11

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.0623. Min Act: -0.0000

Data Index: 1592062 (Python Code)

Max Activating Token Index: 636

Text #12

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.0526. Min Act: -0.0000

Data Index: 525693 (C4 (Web Text))

Max Activating Token Index: 173

Text #13

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.0568. Min Act: -0.0000

Data Index: 1376994 (Python Code)

Max Activating Token Index: 171

Text #14

Max Range: 1.4053. Min Range: -1.4053

Max Act: 0.9224. Min Act: -0.0000

Data Index: 770277 (C4 (Web Text))

Max Activating Token Index: 283

Text #15

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.0448. Min Act: -0.0000

Data Index: 927017 (C4 (Web Text))

Max Activating Token Index: 528

Text #16

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.0501. Min Act: -0.0000

Data Index: 503274 (C4 (Web Text))

Max Activating Token Index: 6

Text #17

Max Range: 1.4053. Min Range: -1.4053

Max Act: 0.9665. Min Act: -0.0000

Data Index: 476117 (C4 (Web Text))

Max Activating Token Index: 600

Text #18

Max Range: 1.4053. Min Range: -1.4053

Max Act: 0.9240. Min Act: -0.0000

Data Index: 1564049 (Python Code)

Max Activating Token Index: 293

Text #19

Max Range: 1.4053. Min Range: -1.4053

Max Act: 1.0167. Min Act: -0.0000

Data Index: 1431788 (Python Code)

Max Activating Token Index: 134