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

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

Neuron 1988 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: 5.6734. Min Range: -5.6734

Max Act: 5.6734. Min Act: -0.1700

Data Index: 1151522 (C4 (Web Text))

Max Activating Token Index: 198

Text #1

Max Range: 5.6734. Min Range: -5.6734

Max Act: 5.0015. Min Act: -0.1700

Data Index: 189181 (C4 (Web Text))

Max Activating Token Index: 274

Text #2

Max Range: 5.6734. Min Range: -5.6734

Max Act: 5.0171. Min Act: -0.1700

Data Index: 679067 (C4 (Web Text))

Max Activating Token Index: 655

Text #3

Max Range: 5.6734. Min Range: -5.6734

Max Act: 5.0015. Min Act: -0.1700

Data Index: 42799 (C4 (Web Text))

Max Activating Token Index: 537

Text #4

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.9546. Min Act: -0.1700

Data Index: 659185 (C4 (Web Text))

Max Activating Token Index: 360

Text #5

Max Range: 5.6734. Min Range: -5.6734

Max Act: 5.0171. Min Act: -0.1700

Data Index: 705544 (C4 (Web Text))

Max Activating Token Index: 383

Text #6

Max Range: 5.6734. Min Range: -5.6734

Max Act: 5.0328. Min Act: -0.1700

Data Index: 997156 (C4 (Web Text))

Max Activating Token Index: 814

Text #7

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.8921. Min Act: -0.1700

Data Index: 541600 (C4 (Web Text))

Max Activating Token Index: 953

Text #8

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.9078. Min Act: -0.1700

Data Index: 113903 (C4 (Web Text))

Max Activating Token Index: 483

Text #9

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.9390. Min Act: -0.1700

Data Index: 14968 (C4 (Web Text))

Max Activating Token Index: 521

Text #10

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.7984. Min Act: -0.1700

Data Index: 328435 (C4 (Web Text))

Max Activating Token Index: 447

Text #11

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.9078. Min Act: -0.1700

Data Index: 402757 (C4 (Web Text))

Max Activating Token Index: 260

Text #12

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.9234. Min Act: -0.1700

Data Index: 76789 (C4 (Web Text))

Max Activating Token Index: 847

Text #13

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.7984. Min Act: -0.1700

Data Index: 985279 (C4 (Web Text))

Max Activating Token Index: 802

Text #14

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.8296. Min Act: -0.1700

Data Index: 227643 (C4 (Web Text))

Max Activating Token Index: 400

Text #15

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.8140. Min Act: -0.1700

Data Index: 309627 (C4 (Web Text))

Max Activating Token Index: 810

Text #16

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.7984. Min Act: -0.1700

Data Index: 267669 (C4 (Web Text))

Max Activating Token Index: 712

Text #17

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.9078. Min Act: -0.1700

Data Index: 713799 (C4 (Web Text))

Max Activating Token Index: 188

Text #18

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.8609. Min Act: -0.1700

Data Index: 399057 (C4 (Web Text))

Max Activating Token Index: 300

Text #19

Max Range: 5.6734. Min Range: -5.6734

Max Act: 4.8453. Min Act: -0.1700

Data Index: 212763 (C4 (Web Text))

Max Activating Token Index: 817