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

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

Neuron 28 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: 5.9810. Min Range: -5.9810

Max Act: 5.9497. Min Act: -0.1700

Data Index: 394065 (C4 (Web Text))

Max Activating Token Index: 995

Text #1

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.9810. Min Act: -0.1700

Data Index: 215207 (C4 (Web Text))

Max Activating Token Index: 1005

Text #2

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.9810. Min Act: -0.1700

Data Index: 302499 (C4 (Web Text))

Max Activating Token Index: 717

Text #3

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.9497. Min Act: -0.1700

Data Index: 399531 (C4 (Web Text))

Max Activating Token Index: 1001

Text #4

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.8560. Min Act: -0.1700

Data Index: 884002 (C4 (Web Text))

Max Activating Token Index: 361

Text #5

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.7935. Min Act: -0.1700

Data Index: 1103724 (C4 (Web Text))

Max Activating Token Index: 91

Text #6

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.7622. Min Act: -0.1700

Data Index: 1180742 (C4 (Web Text))

Max Activating Token Index: 930

Text #7

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6997. Min Act: -0.1700

Data Index: 572152 (C4 (Web Text))

Max Activating Token Index: 877

Text #8

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6685. Min Act: -0.1700

Data Index: 1312149 (C4 (Web Text))

Max Activating Token Index: 647

Text #9

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6372. Min Act: -0.1700

Data Index: 722278 (C4 (Web Text))

Max Activating Token Index: 597

Text #10

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6685. Min Act: -0.1700

Data Index: 765240 (C4 (Web Text))

Max Activating Token Index: 932

Text #11

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6372. Min Act: -0.1700

Data Index: 301182 (C4 (Web Text))

Max Activating Token Index: 72

Text #12

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6685. Min Act: -0.1700

Data Index: 878200 (C4 (Web Text))

Max Activating Token Index: 427

Text #13

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6685. Min Act: -0.1700

Data Index: 906929 (C4 (Web Text))

Max Activating Token Index: 968

Text #14

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6997. Min Act: -0.1700

Data Index: 71456 (C4 (Web Text))

Max Activating Token Index: 922

Text #15

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6685. Min Act: -0.1700

Data Index: 1138745 (C4 (Web Text))

Max Activating Token Index: 315

Text #16

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6685. Min Act: -0.1700

Data Index: 893257 (C4 (Web Text))

Max Activating Token Index: 484

Text #17

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.5747. Min Act: -0.1700

Data Index: 1011744 (C4 (Web Text))

Max Activating Token Index: 645

Text #18

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6372. Min Act: -0.1700

Data Index: 684699 (C4 (Web Text))

Max Activating Token Index: 709

Text #19

Max Range: 5.9810. Min Range: -5.9810

Max Act: 5.6685. Min Act: -0.1700

Data Index: 502034 (C4 (Web Text))

Max Activating Token Index: 39