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

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

Neuron 1400 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.0491. Min Range: -3.0491

Max Act: 3.0491. Min Act: -0.1671

Data Index: 1551897 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.5228. Min Act: -0.1700

Data Index: 73628 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4904. Min Act: -0.1700

Data Index: 668347 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.5066. Min Act: -0.1700

Data Index: 1484158 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4742. Min Act: -0.1700

Data Index: 589114 (C4 (Web Text))

Max Activating Token Index: 986

Text #5

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4580. Min Act: -0.1698

Data Index: 1433814 (Python Code)

Max Activating Token Index: 932

Text #6

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4417. Min Act: -0.1700

Data Index: 1483592 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4417. Min Act: -0.1700

Data Index: 1022690 (C4 (Web Text))

Max Activating Token Index: 1012

Text #8

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4417. Min Act: -0.1700

Data Index: 297098 (C4 (Web Text))

Max Activating Token Index: 866

Text #9

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4417. Min Act: -0.1700

Data Index: 452693 (C4 (Web Text))

Max Activating Token Index: 802

Text #10

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4417. Min Act: -0.1700

Data Index: 227912 (C4 (Web Text))

Max Activating Token Index: 865

Text #11

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4254. Min Act: -0.1700

Data Index: 1536066 (Python Code)

Max Activating Token Index: 895

Text #12

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4417. Min Act: -0.1700

Data Index: 1451780 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4254. Min Act: -0.1700

Data Index: 1505801 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.3928. Min Act: -0.1700

Data Index: 98084 (C4 (Web Text))

Max Activating Token Index: 463

Text #15

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4254. Min Act: -0.1700

Data Index: 1462627 (Python Code)

Max Activating Token Index: 841

Text #16

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4091. Min Act: -0.1699

Data Index: 1493581 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.4254. Min Act: -0.1700

Data Index: 1535343 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.3928. Min Act: -0.1700

Data Index: 173010 (C4 (Web Text))

Max Activating Token Index: 849

Text #19

Max Range: 3.0491. Min Range: -3.0491

Max Act: 2.3928. Min Act: -0.1700

Data Index: 1005972 (C4 (Web Text))

Max Activating Token Index: 443