Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1920 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.0308. Min Range: -4.0308

Max Act: 4.0308. Min Act: -0.0000

Data Index: 120548 (C4 (Web Text))

Max Activating Token Index: 688

Text #1

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.7411. Min Act: -0.0000

Data Index: 1089078 (C4 (Web Text))

Max Activating Token Index: 591

Text #2

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.8931. Min Act: -0.0000

Data Index: 292711 (C4 (Web Text))

Max Activating Token Index: 165

Text #3

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.7895. Min Act: -0.0000

Data Index: 848177 (C4 (Web Text))

Max Activating Token Index: 583

Text #4

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.4414. Min Act: -0.0000

Data Index: 1193063 (C4 (Web Text))

Max Activating Token Index: 299

Text #5

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.6080. Min Act: -0.0000

Data Index: 327204 (C4 (Web Text))

Max Activating Token Index: 142

Text #6

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.5902. Min Act: -0.0000

Data Index: 427753 (C4 (Web Text))

Max Activating Token Index: 599

Text #7

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.5280. Min Act: -0.0000

Data Index: 859983 (C4 (Web Text))

Max Activating Token Index: 250

Text #8

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.6818. Min Act: -0.0000

Data Index: 915069 (C4 (Web Text))

Max Activating Token Index: 301

Text #9

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.4951. Min Act: -0.0000

Data Index: 70374 (C4 (Web Text))

Max Activating Token Index: 394

Text #10

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.4777. Min Act: -0.0000

Data Index: 931626 (C4 (Web Text))

Max Activating Token Index: 405

Text #11

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.4673. Min Act: -0.0000

Data Index: 793634 (C4 (Web Text))

Max Activating Token Index: 651

Text #12

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.4606. Min Act: -0.0000

Data Index: 620219 (C4 (Web Text))

Max Activating Token Index: 527

Text #13

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.5951. Min Act: -0.0000

Data Index: 169381 (C4 (Web Text))

Max Activating Token Index: 660

Text #14

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.5410. Min Act: -0.0000

Data Index: 1626458 (Python Code)

Max Activating Token Index: 134

Text #15

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.4827. Min Act: -0.0000

Data Index: 252257 (C4 (Web Text))

Max Activating Token Index: 148

Text #16

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.3235. Min Act: -0.0000

Data Index: 647773 (C4 (Web Text))

Max Activating Token Index: 271

Text #17

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.3204. Min Act: -0.0000

Data Index: 988924 (C4 (Web Text))

Max Activating Token Index: 713

Text #18

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.3246. Min Act: -0.0000

Data Index: 407820 (C4 (Web Text))

Max Activating Token Index: 569

Text #19

Max Range: 4.0308. Min Range: -4.0308

Max Act: 3.1221. Min Act: -0.0000

Data Index: 727128 (C4 (Web Text))

Max Activating Token Index: 674