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

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

Neuron 2564 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.3854. Min Range: -8.3854

Max Act: 8.3854. Min Act: -0.0001

Data Index: 226115 (C4 (Web Text))

Max Activating Token Index: 192

Text #1

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.6193. Min Act: -0.0001

Data Index: 699545 (C4 (Web Text))

Max Activating Token Index: 249

Text #2

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.6480. Min Act: -0.0001

Data Index: 606945 (C4 (Web Text))

Max Activating Token Index: 504

Text #3

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.6591. Min Act: -0.0001

Data Index: 923709 (C4 (Web Text))

Max Activating Token Index: 762

Text #4

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.4954. Min Act: -0.0001

Data Index: 862127 (C4 (Web Text))

Max Activating Token Index: 150

Text #5

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.6314. Min Act: -0.0001

Data Index: 864823 (C4 (Web Text))

Max Activating Token Index: 213

Text #6

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.4685. Min Act: -0.0001

Data Index: 1232786 (C4 (Web Text))

Max Activating Token Index: 632

Text #7

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.2869. Min Act: -0.0001

Data Index: 876489 (C4 (Web Text))

Max Activating Token Index: 174

Text #8

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.2510. Min Act: -0.0001

Data Index: 94562 (C4 (Web Text))

Max Activating Token Index: 618

Text #9

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.3905. Min Act: -0.0001

Data Index: 67212 (C4 (Web Text))

Max Activating Token Index: 564

Text #10

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.3687. Min Act: -0.0001

Data Index: 465284 (C4 (Web Text))

Max Activating Token Index: 730

Text #11

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.4945. Min Act: -0.0001

Data Index: 424119 (C4 (Web Text))

Max Activating Token Index: 936

Text #12

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.3833. Min Act: -0.0001

Data Index: 1309789 (C4 (Web Text))

Max Activating Token Index: 359

Text #13

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.2157. Min Act: -0.0001

Data Index: 187944 (C4 (Web Text))

Max Activating Token Index: 292

Text #14

Max Range: 8.3854. Min Range: -8.3854

Max Act: 7.1893. Min Act: -0.0001

Data Index: 959043 (C4 (Web Text))

Max Activating Token Index: 713

Text #15

Max Range: 8.3854. Min Range: -8.3854

Max Act: 6.9717. Min Act: -0.0001

Data Index: 1125384 (C4 (Web Text))

Max Activating Token Index: 961

Text #16

Max Range: 8.3854. Min Range: -8.3854

Max Act: 6.9662. Min Act: -0.0001

Data Index: 915132 (C4 (Web Text))

Max Activating Token Index: 125

Text #17

Max Range: 8.3854. Min Range: -8.3854

Max Act: 6.9481. Min Act: -0.0001

Data Index: 13045 (C4 (Web Text))

Max Activating Token Index: 286

Text #18

Max Range: 8.3854. Min Range: -8.3854

Max Act: 6.8924. Min Act: -0.0001

Data Index: 1287911 (C4 (Web Text))

Max Activating Token Index: 843

Text #19

Max Range: 8.3854. Min Range: -8.3854

Max Act: 6.8598. Min Act: -0.0001

Data Index: 1054279 (C4 (Web Text))

Max Activating Token Index: 999