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

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

Neuron 756 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: 5.8764. Min Range: -5.8764

Max Act: 5.8764. Min Act: -0.1700

Data Index: 1030023 (C4 (Web Text))

Max Activating Token Index: 503

Text #1

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.7826. Min Act: -0.1700

Data Index: 1281148 (C4 (Web Text))

Max Activating Token Index: 274

Text #2

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.3764. Min Act: -0.1700

Data Index: 535884 (C4 (Web Text))

Max Activating Token Index: 845

Text #3

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.2826. Min Act: -0.1700

Data Index: 989420 (C4 (Web Text))

Max Activating Token Index: 1011

Text #4

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.3139. Min Act: -0.1700

Data Index: 466120 (C4 (Web Text))

Max Activating Token Index: 966

Text #5

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.2201. Min Act: -0.1700

Data Index: 410291 (C4 (Web Text))

Max Activating Token Index: 597

Text #6

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.1889. Min Act: -0.1700

Data Index: 867773 (C4 (Web Text))

Max Activating Token Index: 95

Text #7

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.1576. Min Act: -0.1700

Data Index: 846316 (C4 (Web Text))

Max Activating Token Index: 149

Text #8

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.1264. Min Act: -0.1700

Data Index: 1352696 (C4 (Web Text))

Max Activating Token Index: 633

Text #9

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.1264. Min Act: -0.1700

Data Index: 290158 (C4 (Web Text))

Max Activating Token Index: 154

Text #10

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.0951. Min Act: -0.1700

Data Index: 566545 (C4 (Web Text))

Max Activating Token Index: 923

Text #11

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.1264. Min Act: -0.1700

Data Index: 201353 (C4 (Web Text))

Max Activating Token Index: 762

Text #12

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.1264. Min Act: -0.1700

Data Index: 602189 (C4 (Web Text))

Max Activating Token Index: 48

Text #13

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.0951. Min Act: -0.1700

Data Index: 937982 (C4 (Web Text))

Max Activating Token Index: 642

Text #14

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.1576. Min Act: -0.1700

Data Index: 804251 (C4 (Web Text))

Max Activating Token Index: 113

Text #15

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.0639. Min Act: -0.1700

Data Index: 824211 (C4 (Web Text))

Max Activating Token Index: 151

Text #16

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.0951. Min Act: -0.1700

Data Index: 748838 (C4 (Web Text))

Max Activating Token Index: 424

Text #17

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.0639. Min Act: -0.1700

Data Index: 1465432 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.0326. Min Act: -0.1700

Data Index: 285240 (C4 (Web Text))

Max Activating Token Index: 96

Text #19

Max Range: 5.8764. Min Range: -5.8764

Max Act: 5.1576. Min Act: -0.1700

Data Index: 172072 (C4 (Web Text))

Max Activating Token Index: 696