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

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

Neuron 476 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: 5.1766. Min Range: -5.1766

Max Act: 5.1766. Min Act: -0.1700

Data Index: 72431 (C4 (Web Text))

Max Activating Token Index: 705

Text #1

Max Range: 5.1766. Min Range: -5.1766

Max Act: 5.0829. Min Act: -0.1700

Data Index: 1017323 (C4 (Web Text))

Max Activating Token Index: 745

Text #2

Max Range: 5.1766. Min Range: -5.1766

Max Act: 5.0204. Min Act: -0.1700

Data Index: 603324 (C4 (Web Text))

Max Activating Token Index: 1011

Text #3

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.9579. Min Act: -0.1700

Data Index: 777239 (C4 (Web Text))

Max Activating Token Index: 821

Text #4

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.9579. Min Act: -0.1700

Data Index: 282953 (C4 (Web Text))

Max Activating Token Index: 425

Text #5

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.9579. Min Act: -0.1700

Data Index: 338350 (C4 (Web Text))

Max Activating Token Index: 790

Text #6

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8954. Min Act: -0.1699

Data Index: 698498 (C4 (Web Text))

Max Activating Token Index: 7

Text #7

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8641. Min Act: -0.1700

Data Index: 724139 (C4 (Web Text))

Max Activating Token Index: 290

Text #8

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8954. Min Act: -0.1700

Data Index: 507760 (C4 (Web Text))

Max Activating Token Index: 89

Text #9

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8328. Min Act: -0.1700

Data Index: 1200106 (C4 (Web Text))

Max Activating Token Index: 75

Text #10

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8641. Min Act: -0.1700

Data Index: 621842 (C4 (Web Text))

Max Activating Token Index: 453

Text #11

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8641. Min Act: -0.1700

Data Index: 772931 (C4 (Web Text))

Max Activating Token Index: 282

Text #12

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8328. Min Act: -0.1700

Data Index: 85717 (C4 (Web Text))

Max Activating Token Index: 291

Text #13

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8641. Min Act: -0.1700

Data Index: 741171 (C4 (Web Text))

Max Activating Token Index: 42

Text #14

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8328. Min Act: -0.1700

Data Index: 1007971 (C4 (Web Text))

Max Activating Token Index: 675

Text #15

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.8016. Min Act: -0.1700

Data Index: 1234418 (C4 (Web Text))

Max Activating Token Index: 369

Text #16

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.7703. Min Act: -0.1700

Data Index: 245062 (C4 (Web Text))

Max Activating Token Index: 716

Text #17

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.7703. Min Act: -0.1700

Data Index: 12267 (C4 (Web Text))

Max Activating Token Index: 161

Text #18

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.7703. Min Act: -0.1700

Data Index: 1052357 (C4 (Web Text))

Max Activating Token Index: 371

Text #19

Max Range: 5.1766. Min Range: -5.1766

Max Act: 4.7391. Min Act: -0.1700

Data Index: 682623 (C4 (Web Text))

Max Activating Token Index: 293