Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1454 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.4890. Min Act: -0.1700

Data Index: 1069543 (C4 (Web Text))

Max Activating Token Index: 799

Text #1

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.3953. Min Act: -0.1700

Data Index: 552186 (C4 (Web Text))

Max Activating Token Index: 423

Text #2

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.3953. Min Act: -0.1700

Data Index: 327092 (C4 (Web Text))

Max Activating Token Index: 110

Text #3

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.4265. Min Act: -0.1700

Data Index: 80718 (C4 (Web Text))

Max Activating Token Index: 686

Text #4

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.4265. Min Act: -0.1700

Data Index: 982753 (C4 (Web Text))

Max Activating Token Index: 643

Text #5

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.3328. Min Act: -0.1700

Data Index: 484879 (C4 (Web Text))

Max Activating Token Index: 894

Text #6

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.3953. Min Act: -0.1700

Data Index: 575117 (C4 (Web Text))

Max Activating Token Index: 306

Text #7

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.3328. Min Act: -0.1700

Data Index: 875756 (C4 (Web Text))

Max Activating Token Index: 290

Text #8

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2703. Min Act: -0.1700

Data Index: 46251 (C4 (Web Text))

Max Activating Token Index: 83

Text #9

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2703. Min Act: -0.1700

Data Index: 846614 (C4 (Web Text))

Max Activating Token Index: 350

Text #10

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2078. Min Act: -0.1700

Data Index: 918505 (C4 (Web Text))

Max Activating Token Index: 295

Text #11

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.0515. Min Act: -0.1700

Data Index: 1006733 (C4 (Web Text))

Max Activating Token Index: 257

Text #12

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.1765. Min Act: -0.1700

Data Index: 684490 (C4 (Web Text))

Max Activating Token Index: 720

Text #13

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2078. Min Act: -0.1700

Data Index: 1076196 (C4 (Web Text))

Max Activating Token Index: 131

Text #14

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2390. Min Act: -0.1700

Data Index: 1346755 (C4 (Web Text))

Max Activating Token Index: 155

Text #15

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2390. Min Act: -0.1700

Data Index: 787844 (C4 (Web Text))

Max Activating Token Index: 504

Text #16

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2390. Min Act: -0.1700

Data Index: 570104 (C4 (Web Text))

Max Activating Token Index: 894

Text #17

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2078. Min Act: -0.1700

Data Index: 1355375 (C4 (Web Text))

Max Activating Token Index: 753

Text #18

Max Range: 5.4890. Min Range: -5.4890

Max Act: 5.2078. Min Act: -0.1700

Data Index: 928614 (C4 (Web Text))

Max Activating Token Index: 1018

Text #19

Max Range: 5.4890. Min Range: -5.4890

Max Act: 4.9578. Min Act: -0.1700

Data Index: 898095 (C4 (Web Text))

Max Activating Token Index: 266