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

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

Neuron 1305 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.5382. Min Act: -0.1700

Data Index: 1310028 (C4 (Web Text))

Max Activating Token Index: 795

Text #1

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.5069. Min Act: -0.1700

Data Index: 344200 (C4 (Web Text))

Max Activating Token Index: 792

Text #2

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.4442. Min Act: -0.1700

Data Index: 149410 (C4 (Web Text))

Max Activating Token Index: 864

Text #3

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3501. Min Act: -0.1700

Data Index: 162055 (C4 (Web Text))

Max Activating Token Index: 977

Text #4

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3815. Min Act: -0.1700

Data Index: 1266093 (C4 (Web Text))

Max Activating Token Index: 862

Text #5

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3501. Min Act: -0.1700

Data Index: 62534 (C4 (Web Text))

Max Activating Token Index: 751

Text #6

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3501. Min Act: -0.1700

Data Index: 1393000 (Python Code)

Max Activating Token Index: 182

Text #7

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3815. Min Act: -0.1700

Data Index: 810844 (C4 (Web Text))

Max Activating Token Index: 950

Text #8

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3815. Min Act: -0.1700

Data Index: 971597 (C4 (Web Text))

Max Activating Token Index: 615

Text #9

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3815. Min Act: -0.1700

Data Index: 1108903 (C4 (Web Text))

Max Activating Token Index: 184

Text #10

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3501. Min Act: -0.1700

Data Index: 116662 (C4 (Web Text))

Max Activating Token Index: 762

Text #11

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.2244. Min Act: -0.1700

Data Index: 48341 (C4 (Web Text))

Max Activating Token Index: 187

Text #12

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.2873. Min Act: -0.1700

Data Index: 828163 (C4 (Web Text))

Max Activating Token Index: 793

Text #13

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.2873. Min Act: -0.1700

Data Index: 521185 (C4 (Web Text))

Max Activating Token Index: 392

Text #14

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.2873. Min Act: -0.1700

Data Index: 58827 (C4 (Web Text))

Max Activating Token Index: 189

Text #15

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.2873. Min Act: -0.1700

Data Index: 1050281 (C4 (Web Text))

Max Activating Token Index: 899

Text #16

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.2244. Min Act: -0.1700

Data Index: 1186630 (C4 (Web Text))

Max Activating Token Index: 821

Text #17

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.3187. Min Act: -0.1700

Data Index: 113016 (C4 (Web Text))

Max Activating Token Index: 933

Text #18

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.2244. Min Act: -0.1700

Data Index: 1392288 (Python Code)

Max Activating Token Index: 802

Text #19

Max Range: 3.5382. Min Range: -3.5382

Max Act: 3.2558. Min Act: -0.1700

Data Index: 572278 (C4 (Web Text))

Max Activating Token Index: 766