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

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

Neuron 582 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3799. Min Range: -5.3799

Max Act: 5.3799. Min Act: -0.1700

Data Index: 808784 (C4 (Web Text))

Max Activating Token Index: 988

Text #1

Max Range: 5.3799. Min Range: -5.3799

Max Act: 5.3487. Min Act: -0.1700

Data Index: 853446 (C4 (Web Text))

Max Activating Token Index: 1001

Text #2

Max Range: 5.3799. Min Range: -5.3799

Max Act: 5.1612. Min Act: -0.1700

Data Index: 1067510 (C4 (Web Text))

Max Activating Token Index: 224

Text #3

Max Range: 5.3799. Min Range: -5.3799

Max Act: 5.1299. Min Act: -0.1700

Data Index: 393683 (C4 (Web Text))

Max Activating Token Index: 587

Text #4

Max Range: 5.3799. Min Range: -5.3799

Max Act: 5.0987. Min Act: -0.1700

Data Index: 578823 (C4 (Web Text))

Max Activating Token Index: 481

Text #5

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.9737. Min Act: -0.1700

Data Index: 710297 (C4 (Web Text))

Max Activating Token Index: 202

Text #6

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.9737. Min Act: -0.1700

Data Index: 400264 (C4 (Web Text))

Max Activating Token Index: 443

Text #7

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.9424. Min Act: -0.1700

Data Index: 634560 (C4 (Web Text))

Max Activating Token Index: 171

Text #8

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.9424. Min Act: -0.1700

Data Index: 764462 (C4 (Web Text))

Max Activating Token Index: 529

Text #9

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8799. Min Act: -0.1700

Data Index: 1145062 (C4 (Web Text))

Max Activating Token Index: 399

Text #10

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8799. Min Act: -0.1700

Data Index: 300205 (C4 (Web Text))

Max Activating Token Index: 866

Text #11

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8799. Min Act: -0.1700

Data Index: 924680 (C4 (Web Text))

Max Activating Token Index: 651

Text #12

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8487. Min Act: -0.1700

Data Index: 645088 (C4 (Web Text))

Max Activating Token Index: 439

Text #13

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8799. Min Act: -0.1700

Data Index: 803625 (C4 (Web Text))

Max Activating Token Index: 646

Text #14

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8487. Min Act: -0.1700

Data Index: 631238 (C4 (Web Text))

Max Activating Token Index: 926

Text #15

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8487. Min Act: -0.1700

Data Index: 100974 (C4 (Web Text))

Max Activating Token Index: 314

Text #16

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8799. Min Act: -0.1700

Data Index: 329925 (C4 (Web Text))

Max Activating Token Index: 756

Text #17

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.8174. Min Act: -0.1700

Data Index: 628235 (C4 (Web Text))

Max Activating Token Index: 458

Text #18

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.7862. Min Act: -0.1700

Data Index: 260268 (C4 (Web Text))

Max Activating Token Index: 542

Text #19

Max Range: 5.3799. Min Range: -5.3799

Max Act: 4.7549. Min Act: -0.1700

Data Index: 542490 (C4 (Web Text))

Max Activating Token Index: 568