Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1067 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.9036. Min Act: -0.1700

Data Index: 1520607 (Python Code)

Max Activating Token Index: 419

Text #1

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.8411. Min Act: -0.1700

Data Index: 1170148 (C4 (Web Text))

Max Activating Token Index: 335

Text #2

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.8411. Min Act: -0.1700

Data Index: 1419627 (Python Code)

Max Activating Token Index: 587

Text #3

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.7473. Min Act: -0.1700

Data Index: 738089 (C4 (Web Text))

Max Activating Token Index: 256

Text #4

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.8098. Min Act: -0.1700

Data Index: 743238 (C4 (Web Text))

Max Activating Token Index: 471

Text #5

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.7473. Min Act: -0.1700

Data Index: 1346950 (C4 (Web Text))

Max Activating Token Index: 217

Text #6

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.7473. Min Act: -0.1700

Data Index: 456736 (C4 (Web Text))

Max Activating Token Index: 457

Text #7

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.7161. Min Act: -0.1700

Data Index: 1174686 (C4 (Web Text))

Max Activating Token Index: 258

Text #8

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.7005. Min Act: -0.1700

Data Index: 196067 (C4 (Web Text))

Max Activating Token Index: 205

Text #9

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6848. Min Act: -0.1700

Data Index: 790939 (C4 (Web Text))

Max Activating Token Index: 301

Text #10

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6848. Min Act: -0.1700

Data Index: 1190713 (C4 (Web Text))

Max Activating Token Index: 405

Text #11

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6692. Min Act: -0.1700

Data Index: 418754 (C4 (Web Text))

Max Activating Token Index: 462

Text #12

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6692. Min Act: -0.1700

Data Index: 964198 (C4 (Web Text))

Max Activating Token Index: 453

Text #13

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6380. Min Act: -0.1700

Data Index: 491359 (C4 (Web Text))

Max Activating Token Index: 520

Text #14

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6536. Min Act: -0.1700

Data Index: 1165102 (C4 (Web Text))

Max Activating Token Index: 529

Text #15

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6848. Min Act: -0.1700

Data Index: 945074 (C4 (Web Text))

Max Activating Token Index: 893

Text #16

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6536. Min Act: -0.1700

Data Index: 949384 (C4 (Web Text))

Max Activating Token Index: 232

Text #17

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6380. Min Act: -0.1699

Data Index: 1443791 (Python Code)

Max Activating Token Index: 376

Text #18

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6536. Min Act: -0.1700

Data Index: 790292 (C4 (Web Text))

Max Activating Token Index: 213

Text #19

Max Range: 5.9036. Min Range: -5.9036

Max Act: 5.6067. Min Act: -0.1700

Data Index: 306966 (C4 (Web Text))

Max Activating Token Index: 815