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

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

Neuron 432 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2944. Min Range: -5.2944

Max Act: 5.2944. Min Act: -0.0000

Data Index: 1041685 (C4 (Web Text))

Max Activating Token Index: 417

Text #1

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.9970. Min Act: -0.0000

Data Index: 475689 (C4 (Web Text))

Max Activating Token Index: 217

Text #2

Max Range: 5.2944. Min Range: -5.2944

Max Act: 5.0173. Min Act: -0.0000

Data Index: 503386 (C4 (Web Text))

Max Activating Token Index: 517

Text #3

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.9385. Min Act: -0.0000

Data Index: 152155 (C4 (Web Text))

Max Activating Token Index: 395

Text #4

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.6433. Min Act: -0.0000

Data Index: 766872 (C4 (Web Text))

Max Activating Token Index: 291

Text #5

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.5834. Min Act: -0.0000

Data Index: 648935 (C4 (Web Text))

Max Activating Token Index: 261

Text #6

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.5969. Min Act: -0.0000

Data Index: 558648 (C4 (Web Text))

Max Activating Token Index: 755

Text #7

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.2407. Min Act: -0.0000

Data Index: 1204658 (C4 (Web Text))

Max Activating Token Index: 305

Text #8

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.1879. Min Act: -0.0000

Data Index: 814278 (C4 (Web Text))

Max Activating Token Index: 311

Text #9

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.3135. Min Act: -0.0000

Data Index: 382254 (C4 (Web Text))

Max Activating Token Index: 343

Text #10

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.8860. Min Act: -0.0000

Data Index: 1317914 (C4 (Web Text))

Max Activating Token Index: 338

Text #11

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.9525. Min Act: -0.0000

Data Index: 113205 (C4 (Web Text))

Max Activating Token Index: 267

Text #12

Max Range: 5.2944. Min Range: -5.2944

Max Act: 4.1101. Min Act: -0.0000

Data Index: 1095999 (C4 (Web Text))

Max Activating Token Index: 491

Text #13

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.7904. Min Act: -0.0000

Data Index: 330953 (C4 (Web Text))

Max Activating Token Index: 708

Text #14

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.7835. Min Act: -0.0000

Data Index: 1376336 (Python Code)

Max Activating Token Index: 236

Text #15

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.7779. Min Act: -0.0000

Data Index: 457102 (C4 (Web Text))

Max Activating Token Index: 1017

Text #16

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.6977. Min Act: -0.0000

Data Index: 721436 (C4 (Web Text))

Max Activating Token Index: 226

Text #17

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.4803. Min Act: -0.0000

Data Index: 1333869 (C4 (Web Text))

Max Activating Token Index: 323

Text #18

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.6795. Min Act: -0.0000

Data Index: 1206137 (C4 (Web Text))

Max Activating Token Index: 665

Text #19

Max Range: 5.2944. Min Range: -5.2944

Max Act: 3.8077. Min Act: -0.0000

Data Index: 20348 (C4 (Web Text))

Max Activating Token Index: 200