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

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

Neuron 996 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1148. Min Range: -5.1148

Max Act: 5.1148. Min Act: -0.1700

Data Index: 717207 (C4 (Web Text))

Max Activating Token Index: 561

Text #1

Max Range: 5.1148. Min Range: -5.1148

Max Act: 5.0523. Min Act: -0.1700

Data Index: 985531 (C4 (Web Text))

Max Activating Token Index: 631

Text #2

Max Range: 5.1148. Min Range: -5.1148

Max Act: 5.0211. Min Act: -0.1700

Data Index: 38383 (C4 (Web Text))

Max Activating Token Index: 535

Text #3

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.9898. Min Act: -0.1700

Data Index: 223151 (C4 (Web Text))

Max Activating Token Index: 722

Text #4

Max Range: 5.1148. Min Range: -5.1148

Max Act: 5.0211. Min Act: -0.1700

Data Index: 539137 (C4 (Web Text))

Max Activating Token Index: 771

Text #5

Max Range: 5.1148. Min Range: -5.1148

Max Act: 5.0523. Min Act: -0.1700

Data Index: 953602 (C4 (Web Text))

Max Activating Token Index: 436

Text #6

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8961. Min Act: -0.1700

Data Index: 214758 (C4 (Web Text))

Max Activating Token Index: 901

Text #7

Max Range: 5.1148. Min Range: -5.1148

Max Act: 5.0211. Min Act: -0.1700

Data Index: 197984 (C4 (Web Text))

Max Activating Token Index: 549

Text #8

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.9273. Min Act: -0.1700

Data Index: 447314 (C4 (Web Text))

Max Activating Token Index: 874

Text #9

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8961. Min Act: -0.1700

Data Index: 1156995 (C4 (Web Text))

Max Activating Token Index: 447

Text #10

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.9273. Min Act: -0.1700

Data Index: 1464145 (Python Code)

Max Activating Token Index: 1001

Text #11

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8961. Min Act: -0.1700

Data Index: 1283358 (C4 (Web Text))

Max Activating Token Index: 566

Text #12

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8336. Min Act: -0.1700

Data Index: 348267 (C4 (Web Text))

Max Activating Token Index: 936

Text #13

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8648. Min Act: -0.1700

Data Index: 1273552 (C4 (Web Text))

Max Activating Token Index: 239

Text #14

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8336. Min Act: -0.1700

Data Index: 167902 (C4 (Web Text))

Max Activating Token Index: 996

Text #15

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8023. Min Act: -0.1700

Data Index: 882583 (C4 (Web Text))

Max Activating Token Index: 459

Text #16

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8023. Min Act: -0.1700

Data Index: 551782 (C4 (Web Text))

Max Activating Token Index: 1008

Text #17

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8336. Min Act: -0.1700

Data Index: 199891 (C4 (Web Text))

Max Activating Token Index: 744

Text #18

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8648. Min Act: -0.1700

Data Index: 50143 (C4 (Web Text))

Max Activating Token Index: 360

Text #19

Max Range: 5.1148. Min Range: -5.1148

Max Act: 4.8336. Min Act: -0.1700

Data Index: 810586 (C4 (Web Text))

Max Activating Token Index: 743