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

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

Neuron 3670 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6165. Min Range: -1.6165

Max Act: 1.6165. Min Act: 0.0000

Data Index: 485986 (C4 (Web Text))

Max Activating Token Index: 642

Text #1

Max Range: 1.6165. Min Range: -1.6165

Max Act: 1.0099. Min Act: 0.0000

Data Index: 136355 (C4 (Web Text))

Max Activating Token Index: 602

Text #2

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.9259. Min Act: -0.0000

Data Index: 1593077 (Python Code)

Max Activating Token Index: 139

Text #3

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.8878. Min Act: 0.0000

Data Index: 1352685 (C4 (Web Text))

Max Activating Token Index: 844

Text #4

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.8014. Min Act: 0.0000

Data Index: 1318731 (C4 (Web Text))

Max Activating Token Index: 476

Text #5

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.8490. Min Act: -0.0000

Data Index: 217443 (C4 (Web Text))

Max Activating Token Index: 106

Text #6

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.8055. Min Act: 0.0000

Data Index: 582221 (C4 (Web Text))

Max Activating Token Index: 190

Text #7

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.8640. Min Act: 0.0000

Data Index: 1334731 (C4 (Web Text))

Max Activating Token Index: 465

Text #8

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.8569. Min Act: -0.0000

Data Index: 1483102 (Python Code)

Max Activating Token Index: 428

Text #9

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7691. Min Act: -0.0000

Data Index: 134935 (C4 (Web Text))

Max Activating Token Index: 541

Text #10

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.8327. Min Act: -0.0000

Data Index: 660950 (C4 (Web Text))

Max Activating Token Index: 407

Text #11

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7584. Min Act: 0.0001

Data Index: 807394 (C4 (Web Text))

Max Activating Token Index: 300

Text #12

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7550. Min Act: -0.0000

Data Index: 718163 (C4 (Web Text))

Max Activating Token Index: 231

Text #13

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7571. Min Act: -0.0000

Data Index: 1404273 (Python Code)

Max Activating Token Index: 64

Text #14

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7562. Min Act: -0.0000

Data Index: 1318292 (C4 (Web Text))

Max Activating Token Index: 850

Text #15

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7051. Min Act: -0.0000

Data Index: 1556964 (Python Code)

Max Activating Token Index: 146

Text #16

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7512. Min Act: 0.0001

Data Index: 261626 (C4 (Web Text))

Max Activating Token Index: 322

Text #17

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7692. Min Act: 0.0000

Data Index: 1088628 (C4 (Web Text))

Max Activating Token Index: 122

Text #18

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7389. Min Act: -0.0000

Data Index: 1203299 (C4 (Web Text))

Max Activating Token Index: 1003

Text #19

Max Range: 1.6165. Min Range: -1.6165

Max Act: 0.7390. Min Act: 0.0001

Data Index: 430263 (C4 (Web Text))

Max Activating Token Index: 439