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

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

Neuron 3174 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.3946. Min Range: -2.3946

Max Act: 2.3946. Min Act: -0.0000

Data Index: 696642 (C4 (Web Text))

Max Activating Token Index: 386

Text #1

Max Range: 2.3946. Min Range: -2.3946

Max Act: 2.2312. Min Act: -0.0000

Data Index: 104587 (C4 (Web Text))

Max Activating Token Index: 860

Text #2

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.9953. Min Act: -0.0000

Data Index: 890703 (C4 (Web Text))

Max Activating Token Index: 189

Text #3

Max Range: 2.3946. Min Range: -2.3946

Max Act: 2.0501. Min Act: -0.0000

Data Index: 789017 (C4 (Web Text))

Max Activating Token Index: 333

Text #4

Max Range: 2.3946. Min Range: -2.3946

Max Act: 2.0235. Min Act: -0.0000

Data Index: 230379 (C4 (Web Text))

Max Activating Token Index: 435

Text #5

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.8825. Min Act: -0.0000

Data Index: 1478240 (Python Code)

Max Activating Token Index: 813

Text #6

Max Range: 2.3946. Min Range: -2.3946

Max Act: 2.0068. Min Act: -0.0000

Data Index: 905277 (C4 (Web Text))

Max Activating Token Index: 724

Text #7

Max Range: 2.3946. Min Range: -2.3946

Max Act: 2.0356. Min Act: -0.0000

Data Index: 343604 (C4 (Web Text))

Max Activating Token Index: 609

Text #8

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.7343. Min Act: -0.0000

Data Index: 1230632 (C4 (Web Text))

Max Activating Token Index: 382

Text #9

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.7626. Min Act: -0.0000

Data Index: 175029 (C4 (Web Text))

Max Activating Token Index: 372

Text #10

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.6404. Min Act: -0.0000

Data Index: 880333 (C4 (Web Text))

Max Activating Token Index: 618

Text #11

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.7365. Min Act: -0.0000

Data Index: 305480 (C4 (Web Text))

Max Activating Token Index: 490

Text #12

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.6144. Min Act: -0.0000

Data Index: 936050 (C4 (Web Text))

Max Activating Token Index: 791

Text #13

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.6929. Min Act: -0.0000

Data Index: 166829 (C4 (Web Text))

Max Activating Token Index: 611

Text #14

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.7729. Min Act: -0.0000

Data Index: 1532750 (Python Code)

Max Activating Token Index: 40

Text #15

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.6805. Min Act: -0.0000

Data Index: 886410 (C4 (Web Text))

Max Activating Token Index: 504

Text #16

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.5769. Min Act: 0.0001

Data Index: 320560 (C4 (Web Text))

Max Activating Token Index: 695

Text #17

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.7549. Min Act: -0.0000

Data Index: 1014419 (C4 (Web Text))

Max Activating Token Index: 263

Text #18

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.6512. Min Act: -0.0000

Data Index: 1289622 (C4 (Web Text))

Max Activating Token Index: 633

Text #19

Max Range: 2.3946. Min Range: -2.3946

Max Act: 1.7260. Min Act: -0.0000

Data Index: 1379420 (Python Code)

Max Activating Token Index: 436