Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 630 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 9.1134. Min Range: -9.1134

Max Act: 9.1134. Min Act: -0.1700

Data Index: 1270924 (C4 (Web Text))

Max Activating Token Index: 497

Text #1

Max Range: 9.1134. Min Range: -9.1134

Max Act: 8.6134. Min Act: -0.1700

Data Index: 1167509 (C4 (Web Text))

Max Activating Token Index: 227

Text #2

Max Range: 9.1134. Min Range: -9.1134

Max Act: 8.5509. Min Act: -0.1700

Data Index: 1006761 (C4 (Web Text))

Max Activating Token Index: 409

Text #3

Max Range: 9.1134. Min Range: -9.1134

Max Act: 8.3009. Min Act: -0.1700

Data Index: 976465 (C4 (Web Text))

Max Activating Token Index: 128

Text #4

Max Range: 9.1134. Min Range: -9.1134

Max Act: 8.0509. Min Act: -0.1700

Data Index: 533035 (C4 (Web Text))

Max Activating Token Index: 407

Text #5

Max Range: 9.1134. Min Range: -9.1134

Max Act: 8.3634. Min Act: -0.1700

Data Index: 1051704 (C4 (Web Text))

Max Activating Token Index: 105

Text #6

Max Range: 9.1134. Min Range: -9.1134

Max Act: 8.2384. Min Act: -0.1700

Data Index: 1314708 (C4 (Web Text))

Max Activating Token Index: 90

Text #7

Max Range: 9.1134. Min Range: -9.1134

Max Act: 8.1759. Min Act: -0.1700

Data Index: 615231 (C4 (Web Text))

Max Activating Token Index: 583

Text #8

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.8634. Min Act: -0.1700

Data Index: 593457 (C4 (Web Text))

Max Activating Token Index: 576

Text #9

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.9259. Min Act: -0.1700

Data Index: 537964 (C4 (Web Text))

Max Activating Token Index: 214

Text #10

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.9259. Min Act: -0.1700

Data Index: 363469 (C4 (Web Text))

Max Activating Token Index: 675

Text #11

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.8634. Min Act: -0.1700

Data Index: 966972 (C4 (Web Text))

Max Activating Token Index: 669

Text #12

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.8634. Min Act: -0.1700

Data Index: 497741 (C4 (Web Text))

Max Activating Token Index: 110

Text #13

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.8634. Min Act: -0.1700

Data Index: 962912 (C4 (Web Text))

Max Activating Token Index: 493

Text #14

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.8009. Min Act: -0.1700

Data Index: 171111 (C4 (Web Text))

Max Activating Token Index: 396

Text #15

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.5509. Min Act: -0.1700

Data Index: 762001 (C4 (Web Text))

Max Activating Token Index: 933

Text #16

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.6759. Min Act: -0.1700

Data Index: 19051 (C4 (Web Text))

Max Activating Token Index: 409

Text #17

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.7071. Min Act: -0.1700

Data Index: 1180936 (C4 (Web Text))

Max Activating Token Index: 598

Text #18

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.9259. Min Act: -0.1700

Data Index: 188182 (C4 (Web Text))

Max Activating Token Index: 709

Text #19

Max Range: 9.1134. Min Range: -9.1134

Max Act: 7.7384. Min Act: -0.1700

Data Index: 697877 (C4 (Web Text))

Max Activating Token Index: 952