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

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

Neuron 1593 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: 5.8768. Min Range: -5.8768

Max Act: 5.8768. Min Act: -0.1700

Data Index: 1239340 (C4 (Web Text))

Max Activating Token Index: 629

Text #1

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.8455. Min Act: -0.1700

Data Index: 1214859 (C4 (Web Text))

Max Activating Token Index: 1016

Text #2

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.8455. Min Act: -0.1700

Data Index: 255476 (C4 (Web Text))

Max Activating Token Index: 633

Text #3

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.7518. Min Act: -0.1700

Data Index: 254037 (C4 (Web Text))

Max Activating Token Index: 768

Text #4

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.7518. Min Act: -0.1700

Data Index: 944023 (C4 (Web Text))

Max Activating Token Index: 590

Text #5

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.5955. Min Act: -0.1700

Data Index: 642707 (C4 (Web Text))

Max Activating Token Index: 596

Text #6

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.6580. Min Act: -0.1700

Data Index: 780244 (C4 (Web Text))

Max Activating Token Index: 676

Text #7

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.6580. Min Act: -0.1700

Data Index: 1095273 (C4 (Web Text))

Max Activating Token Index: 484

Text #8

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.6580. Min Act: -0.1700

Data Index: 143641 (C4 (Web Text))

Max Activating Token Index: 699

Text #9

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.6268. Min Act: -0.1700

Data Index: 786276 (C4 (Web Text))

Max Activating Token Index: 130

Text #10

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.6268. Min Act: -0.1700

Data Index: 1094373 (C4 (Web Text))

Max Activating Token Index: 258

Text #11

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.5643. Min Act: -0.1700

Data Index: 465697 (C4 (Web Text))

Max Activating Token Index: 448

Text #12

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.5643. Min Act: -0.1700

Data Index: 37715 (C4 (Web Text))

Max Activating Token Index: 873

Text #13

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.4705. Min Act: -0.1700

Data Index: 663330 (C4 (Web Text))

Max Activating Token Index: 245

Text #14

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.5018. Min Act: -0.1700

Data Index: 965462 (C4 (Web Text))

Max Activating Token Index: 513

Text #15

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.5643. Min Act: -0.1700

Data Index: 1178283 (C4 (Web Text))

Max Activating Token Index: 656

Text #16

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.4705. Min Act: -0.1700

Data Index: 549774 (C4 (Web Text))

Max Activating Token Index: 489

Text #17

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.5018. Min Act: -0.1700

Data Index: 1002389 (C4 (Web Text))

Max Activating Token Index: 69

Text #18

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.5018. Min Act: -0.1700

Data Index: 957477 (C4 (Web Text))

Max Activating Token Index: 196

Text #19

Max Range: 5.8768. Min Range: -5.8768

Max Act: 5.4705. Min Act: -0.1700

Data Index: 1354172 (C4 (Web Text))

Max Activating Token Index: 545