Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2316 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.8388. Min Act: -0.0001

Data Index: 402781 (C4 (Web Text))

Max Activating Token Index: 33

Text #1

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.6196. Min Act: -0.0001

Data Index: 1346055 (C4 (Web Text))

Max Activating Token Index: 143

Text #2

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.6691. Min Act: -0.0000

Data Index: 807531 (C4 (Web Text))

Max Activating Token Index: 697

Text #3

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5991. Min Act: -0.0001

Data Index: 382350 (C4 (Web Text))

Max Activating Token Index: 848

Text #4

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.6410. Min Act: -0.0001

Data Index: 34688 (C4 (Web Text))

Max Activating Token Index: 729

Text #5

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.6078. Min Act: -0.0001

Data Index: 461245 (C4 (Web Text))

Max Activating Token Index: 46

Text #6

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.6098. Min Act: -0.0001

Data Index: 284206 (C4 (Web Text))

Max Activating Token Index: 67

Text #7

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5289. Min Act: -0.0001

Data Index: 700232 (C4 (Web Text))

Max Activating Token Index: 133

Text #8

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5412. Min Act: -0.0001

Data Index: 868689 (C4 (Web Text))

Max Activating Token Index: 823

Text #9

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5555. Min Act: -0.0001

Data Index: 1449908 (Python Code)

Max Activating Token Index: 148

Text #10

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5268. Min Act: -0.0001

Data Index: 1173365 (C4 (Web Text))

Max Activating Token Index: 677

Text #11

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5003. Min Act: -0.0001

Data Index: 987780 (C4 (Web Text))

Max Activating Token Index: 399

Text #12

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5294. Min Act: -0.0001

Data Index: 919444 (C4 (Web Text))

Max Activating Token Index: 464

Text #13

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5112. Min Act: -0.0001

Data Index: 976200 (C4 (Web Text))

Max Activating Token Index: 28

Text #14

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.4901. Min Act: -0.0001

Data Index: 1087623 (C4 (Web Text))

Max Activating Token Index: 240

Text #15

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.4949. Min Act: -0.0000

Data Index: 893208 (C4 (Web Text))

Max Activating Token Index: 560

Text #16

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.4881. Min Act: -0.0001

Data Index: 1466517 (Python Code)

Max Activating Token Index: 481

Text #17

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.4998. Min Act: -0.0001

Data Index: 1304627 (C4 (Web Text))

Max Activating Token Index: 337

Text #18

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.5140. Min Act: -0.0000

Data Index: 510667 (C4 (Web Text))

Max Activating Token Index: 960

Text #19

Max Range: 0.8388. Min Range: -0.8388

Max Act: 0.4834. Min Act: -0.0001

Data Index: 541909 (C4 (Web Text))

Max Activating Token Index: 731