Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 1995 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 5.3974. Min Range: -5.3974

Max Act: 5.3974. Min Act: -0.1700

Data Index: 1641293 (Python Code)

Max Activating Token Index: 942

Text #1

Max Range: 5.3974. Min Range: -5.3974

Max Act: 5.1487. Min Act: -0.1700

Data Index: 1490474 (Python Code)

Max Activating Token Index: 1000

Text #2

Max Range: 5.3974. Min Range: -5.3974

Max Act: 5.0927. Min Act: -0.1699

Data Index: 1255545 (C4 (Web Text))

Max Activating Token Index: 303

Text #3

Max Range: 5.3974. Min Range: -5.3974

Max Act: 5.0512. Min Act: -0.1666

Data Index: 955128 (C4 (Web Text))

Max Activating Token Index: 408

Text #4

Max Range: 5.3974. Min Range: -5.3974

Max Act: 5.0737. Min Act: -0.1699

Data Index: 1191859 (C4 (Web Text))

Max Activating Token Index: 853

Text #5

Max Range: 5.3974. Min Range: -5.3974

Max Act: 5.0430. Min Act: -0.1690

Data Index: 1363742 (Python Code)

Max Activating Token Index: 734

Text #6

Max Range: 5.3974. Min Range: -5.3974

Max Act: 5.0152. Min Act: -0.1700

Data Index: 389765 (C4 (Web Text))

Max Activating Token Index: 273

Text #7

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.9942. Min Act: -0.1700

Data Index: 178546 (C4 (Web Text))

Max Activating Token Index: 158

Text #8

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.9408. Min Act: -0.1700

Data Index: 1313977 (C4 (Web Text))

Max Activating Token Index: 526

Text #9

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.8589. Min Act: -0.1698

Data Index: 1639999 (Python Code)

Max Activating Token Index: 582

Text #10

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.8708. Min Act: -0.1699

Data Index: 67990 (C4 (Web Text))

Max Activating Token Index: 592

Text #11

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.8728. Min Act: -0.1700

Data Index: 262314 (C4 (Web Text))

Max Activating Token Index: 554

Text #12

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.8508. Min Act: -0.1697

Data Index: 1507173 (Python Code)

Max Activating Token Index: 162

Text #13

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.7611. Min Act: -0.1700

Data Index: 7953 (C4 (Web Text))

Max Activating Token Index: 229

Text #14

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.7765. Min Act: -0.1699

Data Index: 886433 (C4 (Web Text))

Max Activating Token Index: 206

Text #15

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.7616. Min Act: -0.1699

Data Index: 822842 (C4 (Web Text))

Max Activating Token Index: 542

Text #16

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.7735. Min Act: -0.1700

Data Index: 745277 (C4 (Web Text))

Max Activating Token Index: 599

Text #17

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.7390. Min Act: -0.1700

Data Index: 1538578 (Python Code)

Max Activating Token Index: 124

Text #18

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.7402. Min Act: -0.1698

Data Index: 1566664 (Python Code)

Max Activating Token Index: 439

Text #19

Max Range: 5.3974. Min Range: -5.3974

Max Act: 4.7585. Min Act: -0.1700

Data Index: 1478834 (Python Code)

Max Activating Token Index: 589