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

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

Neuron 1989 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: 5.7341. Min Range: -5.7341

Max Act: 5.7341. Min Act: -0.1700

Data Index: 468085 (C4 (Web Text))

Max Activating Token Index: 678

Text #1

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.4216. Min Act: -0.1700

Data Index: 194736 (C4 (Web Text))

Max Activating Token Index: 547

Text #2

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.4216. Min Act: -0.1700

Data Index: 810720 (C4 (Web Text))

Max Activating Token Index: 424

Text #3

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.2654. Min Act: -0.1700

Data Index: 261151 (C4 (Web Text))

Max Activating Token Index: 298

Text #4

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.2029. Min Act: -0.1700

Data Index: 529018 (C4 (Web Text))

Max Activating Token Index: 1021

Text #5

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.1716. Min Act: -0.1700

Data Index: 801919 (C4 (Web Text))

Max Activating Token Index: 315

Text #6

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.0779. Min Act: -0.1700

Data Index: 1131075 (C4 (Web Text))

Max Activating Token Index: 625

Text #7

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.0779. Min Act: -0.1700

Data Index: 1060023 (C4 (Web Text))

Max Activating Token Index: 651

Text #8

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.1404. Min Act: -0.1700

Data Index: 798436 (C4 (Web Text))

Max Activating Token Index: 438

Text #9

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.0779. Min Act: -0.1700

Data Index: 353800 (C4 (Web Text))

Max Activating Token Index: 831

Text #10

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.0779. Min Act: -0.1700

Data Index: 1297865 (C4 (Web Text))

Max Activating Token Index: 312

Text #11

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.0466. Min Act: -0.1700

Data Index: 169832 (C4 (Web Text))

Max Activating Token Index: 841

Text #12

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.0154. Min Act: -0.1700

Data Index: 172926 (C4 (Web Text))

Max Activating Token Index: 129

Text #13

Max Range: 5.7341. Min Range: -5.7341

Max Act: 5.0154. Min Act: -0.1700

Data Index: 892152 (C4 (Web Text))

Max Activating Token Index: 325

Text #14

Max Range: 5.7341. Min Range: -5.7341

Max Act: 4.9529. Min Act: -0.1700

Data Index: 176563 (C4 (Web Text))

Max Activating Token Index: 937

Text #15

Max Range: 5.7341. Min Range: -5.7341

Max Act: 4.9841. Min Act: -0.1700

Data Index: 23689 (C4 (Web Text))

Max Activating Token Index: 940

Text #16

Max Range: 5.7341. Min Range: -5.7341

Max Act: 4.8591. Min Act: -0.1700

Data Index: 1215269 (C4 (Web Text))

Max Activating Token Index: 604

Text #17

Max Range: 5.7341. Min Range: -5.7341

Max Act: 4.9529. Min Act: -0.1700

Data Index: 691364 (C4 (Web Text))

Max Activating Token Index: 349

Text #18

Max Range: 5.7341. Min Range: -5.7341

Max Act: 4.9216. Min Act: -0.1700

Data Index: 117394 (C4 (Web Text))

Max Activating Token Index: 1007

Text #19

Max Range: 5.7341. Min Range: -5.7341

Max Act: 4.9529. Min Act: -0.1700

Data Index: 225894 (C4 (Web Text))

Max Activating Token Index: 318