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

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

Neuron 1231 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: 5.4017. Min Range: -5.4017

Max Act: 5.4017. Min Act: -0.1700

Data Index: 543252 (C4 (Web Text))

Max Activating Token Index: 504

Text #1

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.4017. Min Act: -0.1700

Data Index: 738710 (C4 (Web Text))

Max Activating Token Index: 414

Text #2

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.3705. Min Act: -0.1700

Data Index: 936235 (C4 (Web Text))

Max Activating Token Index: 663

Text #3

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.4017. Min Act: -0.1700

Data Index: 1058560 (C4 (Web Text))

Max Activating Token Index: 597

Text #4

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.3079. Min Act: -0.1700

Data Index: 656183 (C4 (Web Text))

Max Activating Token Index: 728

Text #5

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.2767. Min Act: -0.1700

Data Index: 1222314 (C4 (Web Text))

Max Activating Token Index: 474

Text #6

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.3079. Min Act: -0.1700

Data Index: 1122632 (C4 (Web Text))

Max Activating Token Index: 613

Text #7

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.2454. Min Act: -0.1700

Data Index: 943261 (C4 (Web Text))

Max Activating Token Index: 630

Text #8

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.2454. Min Act: -0.1700

Data Index: 1063407 (C4 (Web Text))

Max Activating Token Index: 774

Text #9

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.2142. Min Act: -0.1700

Data Index: 302051 (C4 (Web Text))

Max Activating Token Index: 616

Text #10

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.2454. Min Act: -0.1700

Data Index: 95129 (C4 (Web Text))

Max Activating Token Index: 17

Text #11

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.2142. Min Act: -0.1700

Data Index: 575854 (C4 (Web Text))

Max Activating Token Index: 540

Text #12

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.2142. Min Act: -0.1700

Data Index: 633364 (C4 (Web Text))

Max Activating Token Index: 445

Text #13

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.1829. Min Act: -0.1700

Data Index: 643514 (C4 (Web Text))

Max Activating Token Index: 496

Text #14

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.1517. Min Act: -0.1700

Data Index: 1364281 (Python Code)

Max Activating Token Index: 35

Text #15

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.1829. Min Act: -0.1700

Data Index: 683549 (C4 (Web Text))

Max Activating Token Index: 786

Text #16

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.1829. Min Act: -0.1700

Data Index: 457656 (C4 (Web Text))

Max Activating Token Index: 465

Text #17

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.1517. Min Act: -0.1700

Data Index: 27315 (C4 (Web Text))

Max Activating Token Index: 639

Text #18

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.1517. Min Act: -0.1700

Data Index: 396602 (C4 (Web Text))

Max Activating Token Index: 444

Text #19

Max Range: 5.4017. Min Range: -5.4017

Max Act: 5.1829. Min Act: -0.1700

Data Index: 506183 (C4 (Web Text))

Max Activating Token Index: 23