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

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

Neuron 837 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: 5.6888. Min Range: -5.6888

Max Act: 5.6888. Min Act: -0.1700

Data Index: 396411 (C4 (Web Text))

Max Activating Token Index: 205

Text #1

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.6575. Min Act: -0.1700

Data Index: 888099 (C4 (Web Text))

Max Activating Token Index: 780

Text #2

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.6575. Min Act: -0.1700

Data Index: 956080 (C4 (Web Text))

Max Activating Token Index: 763

Text #3

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.6263. Min Act: -0.1700

Data Index: 190302 (C4 (Web Text))

Max Activating Token Index: 133

Text #4

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.6575. Min Act: -0.1700

Data Index: 807352 (C4 (Web Text))

Max Activating Token Index: 268

Text #5

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5638. Min Act: -0.1700

Data Index: 1085768 (C4 (Web Text))

Max Activating Token Index: 846

Text #6

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5950. Min Act: -0.1700

Data Index: 260975 (C4 (Web Text))

Max Activating Token Index: 904

Text #7

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5013. Min Act: -0.1700

Data Index: 508564 (C4 (Web Text))

Max Activating Token Index: 908

Text #8

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5013. Min Act: -0.1700

Data Index: 140399 (C4 (Web Text))

Max Activating Token Index: 36

Text #9

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5638. Min Act: -0.1700

Data Index: 488252 (C4 (Web Text))

Max Activating Token Index: 873

Text #10

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5325. Min Act: -0.1700

Data Index: 1117676 (C4 (Web Text))

Max Activating Token Index: 583

Text #11

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5325. Min Act: -0.1700

Data Index: 1109045 (C4 (Web Text))

Max Activating Token Index: 672

Text #12

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5325. Min Act: -0.1700

Data Index: 812293 (C4 (Web Text))

Max Activating Token Index: 630

Text #13

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.4700. Min Act: -0.1700

Data Index: 152582 (C4 (Web Text))

Max Activating Token Index: 324

Text #14

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.4700. Min Act: -0.1700

Data Index: 861718 (C4 (Web Text))

Max Activating Token Index: 219

Text #15

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.4388. Min Act: -0.1700

Data Index: 1241081 (C4 (Web Text))

Max Activating Token Index: 349

Text #16

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.4700. Min Act: -0.1700

Data Index: 848655 (C4 (Web Text))

Max Activating Token Index: 531

Text #17

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5013. Min Act: -0.1700

Data Index: 525092 (C4 (Web Text))

Max Activating Token Index: 299

Text #18

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.4700. Min Act: -0.1700

Data Index: 596418 (C4 (Web Text))

Max Activating Token Index: 736

Text #19

Max Range: 5.6888. Min Range: -5.6888

Max Act: 5.5013. Min Act: -0.1700

Data Index: 1342415 (C4 (Web Text))

Max Activating Token Index: 652