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

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

Neuron 351 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1943. Min Range: -3.1943

Max Act: 3.1943. Min Act: -0.1700

Data Index: 1552085 (Python Code)

Max Activating Token Index: 180

Text #1

Max Range: 3.1943. Min Range: -3.1943

Max Act: 3.1628. Min Act: -0.1700

Data Index: 1443318 (Python Code)

Max Activating Token Index: 123

Text #2

Max Range: 3.1943. Min Range: -3.1943

Max Act: 3.1313. Min Act: -0.1700

Data Index: 1517880 (Python Code)

Max Activating Token Index: 1006

Text #3

Max Range: 3.1943. Min Range: -3.1943

Max Act: 3.0998. Min Act: -0.1700

Data Index: 945178 (C4 (Web Text))

Max Activating Token Index: 707

Text #4

Max Range: 3.1943. Min Range: -3.1943

Max Act: 3.0525. Min Act: -0.1700

Data Index: 1548971 (Python Code)

Max Activating Token Index: 954

Text #5

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.9893. Min Act: -0.1700

Data Index: 1415462 (Python Code)

Max Activating Token Index: 861

Text #6

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.9734. Min Act: -0.1700

Data Index: 1394018 (Python Code)

Max Activating Token Index: 168

Text #7

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.9418. Min Act: -0.1700

Data Index: 1364913 (Python Code)

Max Activating Token Index: 26

Text #8

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.9101. Min Act: -0.1700

Data Index: 1367626 (Python Code)

Max Activating Token Index: 841

Text #9

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.9101. Min Act: -0.1700

Data Index: 1384435 (Python Code)

Max Activating Token Index: 1020

Text #10

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8942. Min Act: -0.1700

Data Index: 1307775 (C4 (Web Text))

Max Activating Token Index: 443

Text #11

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8942. Min Act: -0.1700

Data Index: 424098 (C4 (Web Text))

Max Activating Token Index: 651

Text #12

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8783. Min Act: -0.1700

Data Index: 1382361 (Python Code)

Max Activating Token Index: 71

Text #13

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8783. Min Act: -0.1700

Data Index: 1418296 (Python Code)

Max Activating Token Index: 58

Text #14

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8783. Min Act: -0.1700

Data Index: 574201 (C4 (Web Text))

Max Activating Token Index: 52

Text #15

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8466. Min Act: -0.1700

Data Index: 1644398 (Python Code)

Max Activating Token Index: 840

Text #16

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8466. Min Act: -0.1700

Data Index: 1582563 (Python Code)

Max Activating Token Index: 207

Text #17

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8625. Min Act: -0.1700

Data Index: 1490927 (Python Code)

Max Activating Token Index: 513

Text #18

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8625. Min Act: -0.1700

Data Index: 1372401 (Python Code)

Max Activating Token Index: 232

Text #19

Max Range: 3.1943. Min Range: -3.1943

Max Act: 2.8466. Min Act: -0.1700

Data Index: 1656402 (Python Code)

Max Activating Token Index: 78