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

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

Neuron 1578 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: 3.1992. Min Range: -3.1992

Max Act: 3.1992. Min Act: -0.1700

Data Index: 1505738 (Python Code)

Max Activating Token Index: 251

Text #1

Max Range: 3.1992. Min Range: -3.1992

Max Act: 3.1738. Min Act: -0.1700

Data Index: 1426091 (Python Code)

Max Activating Token Index: 24

Text #2

Max Range: 3.1992. Min Range: -3.1992

Max Act: 3.0935. Min Act: -0.1700

Data Index: 1254220 (C4 (Web Text))

Max Activating Token Index: 122

Text #3

Max Range: 3.1992. Min Range: -3.1992

Max Act: 3.0555. Min Act: -0.1700

Data Index: 1282289 (C4 (Web Text))

Max Activating Token Index: 856

Text #4

Max Range: 3.1992. Min Range: -3.1992

Max Act: 3.0321. Min Act: -0.1700

Data Index: 978567 (C4 (Web Text))

Max Activating Token Index: 1004

Text #5

Max Range: 3.1992. Min Range: -3.1992

Max Act: 3.0095. Min Act: -0.1700

Data Index: 604201 (C4 (Web Text))

Max Activating Token Index: 845

Text #6

Max Range: 3.1992. Min Range: -3.1992

Max Act: 3.0039. Min Act: -0.1700

Data Index: 174325 (C4 (Web Text))

Max Activating Token Index: 657

Text #7

Max Range: 3.1992. Min Range: -3.1992

Max Act: 3.0038. Min Act: -0.1700

Data Index: 1583313 (Python Code)

Max Activating Token Index: 902

Text #8

Max Range: 3.1992. Min Range: -3.1992

Max Act: 3.0000. Min Act: -0.1700

Data Index: 1554182 (Python Code)

Max Activating Token Index: 743

Text #9

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9639. Min Act: -0.1700

Data Index: 100860 (C4 (Web Text))

Max Activating Token Index: 268

Text #10

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9612. Min Act: -0.1700

Data Index: 1378362 (Python Code)

Max Activating Token Index: 320

Text #11

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9627. Min Act: -0.1700

Data Index: 1590731 (Python Code)

Max Activating Token Index: 41

Text #12

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9564. Min Act: -0.1700

Data Index: 1626989 (Python Code)

Max Activating Token Index: 453

Text #13

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9523. Min Act: -0.1700

Data Index: 1653689 (Python Code)

Max Activating Token Index: 148

Text #14

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9430. Min Act: -0.1700

Data Index: 957090 (C4 (Web Text))

Max Activating Token Index: 642

Text #15

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9415. Min Act: -0.1700

Data Index: 1458087 (Python Code)

Max Activating Token Index: 658

Text #16

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9302. Min Act: -0.1700

Data Index: 1297225 (C4 (Web Text))

Max Activating Token Index: 458

Text #17

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9271. Min Act: -0.1700

Data Index: 1392269 (Python Code)

Max Activating Token Index: 972

Text #18

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9135. Min Act: -0.1700

Data Index: 1652459 (Python Code)

Max Activating Token Index: 400

Text #19

Max Range: 3.1992. Min Range: -3.1992

Max Act: 2.9132. Min Act: -0.1700

Data Index: 759423 (C4 (Web Text))

Max Activating Token Index: 952