Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1955 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.3853. Min Act: -0.1700

Data Index: 432092 (C4 (Web Text))

Max Activating Token Index: 307

Text #1

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.3228. Min Act: -0.1700

Data Index: 918800 (C4 (Web Text))

Max Activating Token Index: 957

Text #2

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.3228. Min Act: -0.1700

Data Index: 920603 (C4 (Web Text))

Max Activating Token Index: 762

Text #3

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2916. Min Act: -0.1700

Data Index: 402304 (C4 (Web Text))

Max Activating Token Index: 836

Text #4

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.3228. Min Act: -0.1700

Data Index: 709347 (C4 (Web Text))

Max Activating Token Index: 998

Text #5

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.3541. Min Act: -0.1700

Data Index: 1250796 (C4 (Web Text))

Max Activating Token Index: 509

Text #6

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.3541. Min Act: -0.1700

Data Index: 1582506 (Python Code)

Max Activating Token Index: 12

Text #7

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2916. Min Act: -0.1700

Data Index: 792892 (C4 (Web Text))

Max Activating Token Index: 1002

Text #8

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2916. Min Act: -0.1700

Data Index: 1049265 (C4 (Web Text))

Max Activating Token Index: 957

Text #9

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2916. Min Act: -0.1700

Data Index: 314178 (C4 (Web Text))

Max Activating Token Index: 84

Text #10

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2603. Min Act: -0.1700

Data Index: 1189813 (C4 (Web Text))

Max Activating Token Index: 295

Text #11

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.1978. Min Act: -0.1700

Data Index: 222411 (C4 (Web Text))

Max Activating Token Index: 490

Text #12

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2916. Min Act: -0.1700

Data Index: 57263 (C4 (Web Text))

Max Activating Token Index: 406

Text #13

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2916. Min Act: -0.1700

Data Index: 576625 (C4 (Web Text))

Max Activating Token Index: 942

Text #14

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2916. Min Act: -0.1700

Data Index: 980237 (C4 (Web Text))

Max Activating Token Index: 777

Text #15

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2291. Min Act: -0.1700

Data Index: 618246 (C4 (Web Text))

Max Activating Token Index: 29

Text #16

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2603. Min Act: -0.1700

Data Index: 140314 (C4 (Web Text))

Max Activating Token Index: 243

Text #17

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2916. Min Act: -0.1700

Data Index: 248666 (C4 (Web Text))

Max Activating Token Index: 673

Text #18

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.2291. Min Act: -0.1700

Data Index: 1122055 (C4 (Web Text))

Max Activating Token Index: 418

Text #19

Max Range: 5.3853. Min Range: -5.3853

Max Act: 5.1978. Min Act: -0.1698

Data Index: 1157264 (C4 (Web Text))

Max Activating Token Index: 388