Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 1036 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.9789. Min Act: -0.1021

Data Index: 1361419 (Python Code)

Max Activating Token Index: 946

Text #1

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.9221. Min Act: -0.1700

Data Index: 1044374 (C4 (Web Text))

Max Activating Token Index: 91

Text #2

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.9248. Min Act: -0.1700

Data Index: 1157906 (C4 (Web Text))

Max Activating Token Index: 556

Text #3

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.9119. Min Act: -0.1675

Data Index: 1495663 (Python Code)

Max Activating Token Index: 111

Text #4

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8723. Min Act: -0.1700

Data Index: 1149805 (C4 (Web Text))

Max Activating Token Index: 332

Text #5

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8704. Min Act: -0.1700

Data Index: 440491 (C4 (Web Text))

Max Activating Token Index: 1003

Text #6

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8462. Min Act: -0.1695

Data Index: 1442492 (Python Code)

Max Activating Token Index: 387

Text #7

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8256. Min Act: -0.1700

Data Index: 521469 (C4 (Web Text))

Max Activating Token Index: 60

Text #8

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8266. Min Act: -0.1700

Data Index: 1114761 (C4 (Web Text))

Max Activating Token Index: 375

Text #9

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8168. Min Act: -0.1700

Data Index: 1186517 (C4 (Web Text))

Max Activating Token Index: 507

Text #10

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8150. Min Act: -0.1700

Data Index: 712008 (C4 (Web Text))

Max Activating Token Index: 679

Text #11

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8352. Min Act: -0.1700

Data Index: 924266 (C4 (Web Text))

Max Activating Token Index: 993

Text #12

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.8002. Min Act: -0.1700

Data Index: 1254641 (C4 (Web Text))

Max Activating Token Index: 338

Text #13

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.7891. Min Act: -0.1700

Data Index: 801727 (C4 (Web Text))

Max Activating Token Index: 703

Text #14

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.7792. Min Act: -0.1700

Data Index: 646532 (C4 (Web Text))

Max Activating Token Index: 899

Text #15

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.7695. Min Act: -0.1700

Data Index: 200721 (C4 (Web Text))

Max Activating Token Index: 293

Text #16

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.7669. Min Act: -0.1700

Data Index: 738389 (C4 (Web Text))

Max Activating Token Index: 888

Text #17

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.7785. Min Act: -0.1700

Data Index: 641380 (C4 (Web Text))

Max Activating Token Index: 48

Text #18

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.7738. Min Act: -0.1700

Data Index: 1044171 (C4 (Web Text))

Max Activating Token Index: 435

Text #19

Max Range: 2.9789. Min Range: -2.9789

Max Act: 2.7742. Min Act: -0.1700

Data Index: 1223285 (C4 (Web Text))

Max Activating Token Index: 501