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

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

Neuron 3072 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4530. Min Act: -0.0000

Data Index: 1344597 (C4 (Web Text))

Max Activating Token Index: 920

Text #1

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4486. Min Act: -0.0000

Data Index: 776726 (C4 (Web Text))

Max Activating Token Index: 12

Text #2

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4200. Min Act: -0.0000

Data Index: 1062193 (C4 (Web Text))

Max Activating Token Index: 200

Text #3

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4172. Min Act: -0.0000

Data Index: 663502 (C4 (Web Text))

Max Activating Token Index: 645

Text #4

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4146. Min Act: -0.0000

Data Index: 242075 (C4 (Web Text))

Max Activating Token Index: 761

Text #5

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4402. Min Act: -0.0000

Data Index: 931940 (C4 (Web Text))

Max Activating Token Index: 195

Text #6

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4560. Min Act: -0.0000

Data Index: 7060 (C4 (Web Text))

Max Activating Token Index: 768

Text #7

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4075. Min Act: -0.0000

Data Index: 839841 (C4 (Web Text))

Max Activating Token Index: 176

Text #8

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4175. Min Act: -0.0000

Data Index: 840975 (C4 (Web Text))

Max Activating Token Index: 509

Text #9

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4029. Min Act: -0.0000

Data Index: 873537 (C4 (Web Text))

Max Activating Token Index: 967

Text #10

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3975. Min Act: -0.0000

Data Index: 1011190 (C4 (Web Text))

Max Activating Token Index: 242

Text #11

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3942. Min Act: -0.0000

Data Index: 1049403 (C4 (Web Text))

Max Activating Token Index: 35

Text #12

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3825. Min Act: -0.0000

Data Index: 956583 (C4 (Web Text))

Max Activating Token Index: 719

Text #13

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3807. Min Act: -0.0000

Data Index: 938076 (C4 (Web Text))

Max Activating Token Index: 938

Text #14

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3922. Min Act: -0.0000

Data Index: 250228 (C4 (Web Text))

Max Activating Token Index: 25

Text #15

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.4012. Min Act: -0.0000

Data Index: 510326 (C4 (Web Text))

Max Activating Token Index: 678

Text #16

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3740. Min Act: -0.0000

Data Index: 891331 (C4 (Web Text))

Max Activating Token Index: 695

Text #17

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3869. Min Act: -0.0000

Data Index: 217262 (C4 (Web Text))

Max Activating Token Index: 950

Text #18

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3743. Min Act: -0.0000

Data Index: 21070 (C4 (Web Text))

Max Activating Token Index: 13

Text #19

Max Range: 0.4560. Min Range: -0.4560

Max Act: 0.3974. Min Act: -0.0000

Data Index: 895980 (C4 (Web Text))

Max Activating Token Index: 122