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

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

Neuron 1825 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: 0.4091. Min Range: -0.4091

Max Act: 0.4091. Min Act: -0.0000

Data Index: 14511 (C4 (Web Text))

Max Activating Token Index: 243

Text #1

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.3422. Min Act: -0.0000

Data Index: 747588 (C4 (Web Text))

Max Activating Token Index: 208

Text #2

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.3549. Min Act: -0.0000

Data Index: 1101853 (C4 (Web Text))

Max Activating Token Index: 524

Text #3

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.3446. Min Act: -0.0000

Data Index: 46069 (C4 (Web Text))

Max Activating Token Index: 53

Text #4

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.3262. Min Act: -0.0000

Data Index: 1174420 (C4 (Web Text))

Max Activating Token Index: 743

Text #5

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2809. Min Act: -0.0000

Data Index: 215915 (C4 (Web Text))

Max Activating Token Index: 819

Text #6

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2813. Min Act: -0.0000

Data Index: 43061 (C4 (Web Text))

Max Activating Token Index: 902

Text #7

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2782. Min Act: -0.0000

Data Index: 320990 (C4 (Web Text))

Max Activating Token Index: 11

Text #8

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2850. Min Act: -0.0000

Data Index: 492038 (C4 (Web Text))

Max Activating Token Index: 192

Text #9

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2791. Min Act: -0.0000

Data Index: 1510498 (Python Code)

Max Activating Token Index: 115

Text #10

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2699. Min Act: -0.0000

Data Index: 239450 (C4 (Web Text))

Max Activating Token Index: 924

Text #11

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2555. Min Act: -0.0000

Data Index: 70850 (C4 (Web Text))

Max Activating Token Index: 980

Text #12

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2666. Min Act: -0.0000

Data Index: 196675 (C4 (Web Text))

Max Activating Token Index: 943

Text #13

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2740. Min Act: -0.0000

Data Index: 328510 (C4 (Web Text))

Max Activating Token Index: 977

Text #14

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2578. Min Act: -0.0000

Data Index: 1348103 (C4 (Web Text))

Max Activating Token Index: 812

Text #15

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2514. Min Act: -0.0000

Data Index: 62617 (C4 (Web Text))

Max Activating Token Index: 202

Text #16

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2531. Min Act: -0.0000

Data Index: 1048011 (C4 (Web Text))

Max Activating Token Index: 105

Text #17

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2737. Min Act: -0.0000

Data Index: 1648084 (Python Code)

Max Activating Token Index: 682

Text #18

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2469. Min Act: -0.0000

Data Index: 1130025 (C4 (Web Text))

Max Activating Token Index: 743

Text #19

Max Range: 0.4091. Min Range: -0.4091

Max Act: 0.2530. Min Act: -0.0000

Data Index: 1215858 (C4 (Web Text))

Max Activating Token Index: 13