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

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

Neuron 963 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.5452. Min Act: -0.1700

Data Index: 1580442 (Python Code)

Max Activating Token Index: 765

Text #1

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4825. Min Act: -0.1700

Data Index: 778092 (C4 (Web Text))

Max Activating Token Index: 987

Text #2

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4825. Min Act: -0.1700

Data Index: 1539510 (Python Code)

Max Activating Token Index: 218

Text #3

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4825. Min Act: -0.1700

Data Index: 1223545 (C4 (Web Text))

Max Activating Token Index: 286

Text #4

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4512. Min Act: -0.1700

Data Index: 812615 (C4 (Web Text))

Max Activating Token Index: 983

Text #5

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4512. Min Act: -0.1700

Data Index: 1506413 (Python Code)

Max Activating Token Index: 242

Text #6

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4512. Min Act: -0.1700

Data Index: 76577 (C4 (Web Text))

Max Activating Token Index: 115

Text #7

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4825. Min Act: -0.1700

Data Index: 331051 (C4 (Web Text))

Max Activating Token Index: 142

Text #8

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.3885. Min Act: -0.1700

Data Index: 754456 (C4 (Web Text))

Max Activating Token Index: 1014

Text #9

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4198. Min Act: -0.1700

Data Index: 1552196 (Python Code)

Max Activating Token Index: 84

Text #10

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4198. Min Act: -0.1700

Data Index: 1621534 (Python Code)

Max Activating Token Index: 449

Text #11

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4198. Min Act: -0.1700

Data Index: 411932 (C4 (Web Text))

Max Activating Token Index: 167

Text #12

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4198. Min Act: -0.1700

Data Index: 623098 (C4 (Web Text))

Max Activating Token Index: 97

Text #13

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4512. Min Act: -0.1700

Data Index: 250282 (C4 (Web Text))

Max Activating Token Index: 588

Text #14

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.3885. Min Act: -0.1700

Data Index: 1264257 (C4 (Web Text))

Max Activating Token Index: 875

Text #15

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.3571. Min Act: -0.1700

Data Index: 1421110 (Python Code)

Max Activating Token Index: 841

Text #16

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.4198. Min Act: -0.1700

Data Index: 1433180 (Python Code)

Max Activating Token Index: 402

Text #17

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.3885. Min Act: -0.1700

Data Index: 1433869 (Python Code)

Max Activating Token Index: 323

Text #18

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.3571. Min Act: -0.1700

Data Index: 150705 (C4 (Web Text))

Max Activating Token Index: 510

Text #19

Max Range: 3.5452. Min Range: -3.5452

Max Act: 3.3885. Min Act: -0.1700

Data Index: 947745 (C4 (Web Text))

Max Activating Token Index: 89