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

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

Neuron 75 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: 5.7978. Min Range: -5.7978

Max Act: 5.7978. Min Act: -0.1700

Data Index: 290086 (C4 (Web Text))

Max Activating Token Index: 590

Text #1

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.5478. Min Act: -0.1700

Data Index: 1340521 (C4 (Web Text))

Max Activating Token Index: 353

Text #2

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.5790. Min Act: -0.1700

Data Index: 360928 (C4 (Web Text))

Max Activating Token Index: 518

Text #3

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.5165. Min Act: -0.1700

Data Index: 343541 (C4 (Web Text))

Max Activating Token Index: 165

Text #4

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.4853. Min Act: -0.1700

Data Index: 1192416 (C4 (Web Text))

Max Activating Token Index: 291

Text #5

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.4540. Min Act: -0.1700

Data Index: 376309 (C4 (Web Text))

Max Activating Token Index: 749

Text #6

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.4540. Min Act: -0.1700

Data Index: 518059 (C4 (Web Text))

Max Activating Token Index: 286

Text #7

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.4228. Min Act: -0.1700

Data Index: 824697 (C4 (Web Text))

Max Activating Token Index: 166

Text #8

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.4540. Min Act: -0.1700

Data Index: 94955 (C4 (Web Text))

Max Activating Token Index: 632

Text #9

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.3915. Min Act: -0.1700

Data Index: 415192 (C4 (Web Text))

Max Activating Token Index: 340

Text #10

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.4228. Min Act: -0.1700

Data Index: 578095 (C4 (Web Text))

Max Activating Token Index: 318

Text #11

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.4540. Min Act: -0.1700

Data Index: 52863 (C4 (Web Text))

Max Activating Token Index: 579

Text #12

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.3603. Min Act: -0.1700

Data Index: 874956 (C4 (Web Text))

Max Activating Token Index: 864

Text #13

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.4228. Min Act: -0.1700

Data Index: 261875 (C4 (Web Text))

Max Activating Token Index: 187

Text #14

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.3603. Min Act: -0.1700

Data Index: 50421 (C4 (Web Text))

Max Activating Token Index: 148

Text #15

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.3603. Min Act: -0.1700

Data Index: 659843 (C4 (Web Text))

Max Activating Token Index: 841

Text #16

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.3603. Min Act: -0.1700

Data Index: 102827 (C4 (Web Text))

Max Activating Token Index: 790

Text #17

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.3603. Min Act: -0.1700

Data Index: 971723 (C4 (Web Text))

Max Activating Token Index: 916

Text #18

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.2665. Min Act: -0.1700

Data Index: 407105 (C4 (Web Text))

Max Activating Token Index: 1014

Text #19

Max Range: 5.7978. Min Range: -5.7978

Max Act: 5.2978. Min Act: -0.1700

Data Index: 1349748 (C4 (Web Text))

Max Activating Token Index: 737