Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1575 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.3550. Min Range: -5.3550

Max Act: 5.3550. Min Act: -0.1700

Data Index: 1045785 (C4 (Web Text))

Max Activating Token Index: 611

Text #1

Max Range: 5.3550. Min Range: -5.3550

Max Act: 5.0582. Min Act: -0.1700

Data Index: 268508 (C4 (Web Text))

Max Activating Token Index: 886

Text #2

Max Range: 5.3550. Min Range: -5.3550

Max Act: 5.0582. Min Act: -0.1700

Data Index: 1352399 (C4 (Web Text))

Max Activating Token Index: 571

Text #3

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.9332. Min Act: -0.1700

Data Index: 748630 (C4 (Web Text))

Max Activating Token Index: 142

Text #4

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.8707. Min Act: -0.1700

Data Index: 1041930 (C4 (Web Text))

Max Activating Token Index: 540

Text #5

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.8550. Min Act: -0.1700

Data Index: 27431 (C4 (Web Text))

Max Activating Token Index: 553

Text #6

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.6363. Min Act: -0.1700

Data Index: 371324 (C4 (Web Text))

Max Activating Token Index: 456

Text #7

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.7457. Min Act: -0.1700

Data Index: 955408 (C4 (Web Text))

Max Activating Token Index: 674

Text #8

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.7144. Min Act: -0.1700

Data Index: 733332 (C4 (Web Text))

Max Activating Token Index: 347

Text #9

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.7769. Min Act: -0.1700

Data Index: 27647 (C4 (Web Text))

Max Activating Token Index: 693

Text #10

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.6832. Min Act: -0.1700

Data Index: 926538 (C4 (Web Text))

Max Activating Token Index: 464

Text #11

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.6519. Min Act: -0.1700

Data Index: 1340458 (C4 (Web Text))

Max Activating Token Index: 674

Text #12

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.5582. Min Act: -0.1700

Data Index: 657003 (C4 (Web Text))

Max Activating Token Index: 728

Text #13

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.6363. Min Act: -0.1700

Data Index: 1139265 (C4 (Web Text))

Max Activating Token Index: 325

Text #14

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.6519. Min Act: -0.1700

Data Index: 480610 (C4 (Web Text))

Max Activating Token Index: 944

Text #15

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.6050. Min Act: -0.1700

Data Index: 544637 (C4 (Web Text))

Max Activating Token Index: 497

Text #16

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.5894. Min Act: -0.1700

Data Index: 721210 (C4 (Web Text))

Max Activating Token Index: 445

Text #17

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.6207. Min Act: -0.1700

Data Index: 167318 (C4 (Web Text))

Max Activating Token Index: 74

Text #18

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.6675. Min Act: -0.1700

Data Index: 825400 (C4 (Web Text))

Max Activating Token Index: 283

Text #19

Max Range: 5.3550. Min Range: -5.3550

Max Act: 4.5894. Min Act: -0.1700

Data Index: 5173 (C4 (Web Text))

Max Activating Token Index: 599