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

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

Neuron 481 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.0088. Min Range: -4.0088

Max Act: 4.0088. Min Act: -0.1700

Data Index: 1099171 (C4 (Web Text))

Max Activating Token Index: 357

Text #1

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.9775. Min Act: -0.1700

Data Index: 992088 (C4 (Web Text))

Max Activating Token Index: 257

Text #2

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.9463. Min Act: -0.1700

Data Index: 1136185 (C4 (Web Text))

Max Activating Token Index: 675

Text #3

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.8837. Min Act: -0.1700

Data Index: 1280055 (C4 (Web Text))

Max Activating Token Index: 974

Text #4

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.8837. Min Act: -0.1700

Data Index: 1092448 (C4 (Web Text))

Max Activating Token Index: 753

Text #5

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7899. Min Act: -0.1700

Data Index: 860811 (C4 (Web Text))

Max Activating Token Index: 988

Text #6

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.8055. Min Act: -0.1700

Data Index: 1030133 (C4 (Web Text))

Max Activating Token Index: 354

Text #7

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7899. Min Act: -0.1700

Data Index: 537478 (C4 (Web Text))

Max Activating Token Index: 809

Text #8

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.8368. Min Act: -0.1700

Data Index: 307907 (C4 (Web Text))

Max Activating Token Index: 982

Text #9

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.9150. Min Act: -0.1700

Data Index: 592519 (C4 (Web Text))

Max Activating Token Index: 587

Text #10

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.8212. Min Act: -0.1700

Data Index: 732541 (C4 (Web Text))

Max Activating Token Index: 797

Text #11

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.8055. Min Act: -0.1700

Data Index: 1296085 (C4 (Web Text))

Max Activating Token Index: 365

Text #12

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.8055. Min Act: -0.1700

Data Index: 1036508 (C4 (Web Text))

Max Activating Token Index: 944

Text #13

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7429. Min Act: -0.1700

Data Index: 1033371 (C4 (Web Text))

Max Activating Token Index: 558

Text #14

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7742. Min Act: -0.1700

Data Index: 297849 (C4 (Web Text))

Max Activating Token Index: 816

Text #15

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7586. Min Act: -0.1700

Data Index: 688006 (C4 (Web Text))

Max Activating Token Index: 639

Text #16

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7899. Min Act: -0.1700

Data Index: 1206803 (C4 (Web Text))

Max Activating Token Index: 350

Text #17

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7899. Min Act: -0.1700

Data Index: 268914 (C4 (Web Text))

Max Activating Token Index: 399

Text #18

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7586. Min Act: -0.1700

Data Index: 980237 (C4 (Web Text))

Max Activating Token Index: 802

Text #19

Max Range: 4.0088. Min Range: -4.0088

Max Act: 3.7429. Min Act: -0.1700

Data Index: 124950 (C4 (Web Text))

Max Activating Token Index: 312