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

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

Neuron 1110 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: 4.4578. Min Range: -4.4578

Max Act: 4.4578. Min Act: -0.1688

Data Index: 1622063 (Python Code)

Max Activating Token Index: 136

Text #1

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.3559. Min Act: -0.1622

Data Index: 1579080 (Python Code)

Max Activating Token Index: 258

Text #2

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.3528. Min Act: -0.1697

Data Index: 1645951 (Python Code)

Max Activating Token Index: 442

Text #3

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.2868. Min Act: -0.1700

Data Index: 1560977 (Python Code)

Max Activating Token Index: 95

Text #4

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.2862. Min Act: -0.1700

Data Index: 1595440 (Python Code)

Max Activating Token Index: 314

Text #5

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.2240. Min Act: -0.1700

Data Index: 1433412 (Python Code)

Max Activating Token Index: 665

Text #6

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.1678. Min Act: -0.1700

Data Index: 1542108 (Python Code)

Max Activating Token Index: 367

Text #7

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.1427. Min Act: -0.1686

Data Index: 1522381 (Python Code)

Max Activating Token Index: 716

Text #8

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.1080. Min Act: -0.1699

Data Index: 1469875 (Python Code)

Max Activating Token Index: 249

Text #9

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.0568. Min Act: -0.1693

Data Index: 1651386 (Python Code)

Max Activating Token Index: 461

Text #10

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.0340. Min Act: -0.1700

Data Index: 1645032 (Python Code)

Max Activating Token Index: 589

Text #11

Max Range: 4.4578. Min Range: -4.4578

Max Act: 4.0141. Min Act: -0.1700

Data Index: 624638 (C4 (Web Text))

Max Activating Token Index: 675

Text #12

Max Range: 4.4578. Min Range: -4.4578

Max Act: 3.9944. Min Act: -0.1700

Data Index: 1039889 (C4 (Web Text))

Max Activating Token Index: 932

Text #13

Max Range: 4.4578. Min Range: -4.4578

Max Act: 3.9963. Min Act: -0.1700

Data Index: 1519287 (Python Code)

Max Activating Token Index: 901

Text #14

Max Range: 4.4578. Min Range: -4.4578

Max Act: 3.9780. Min Act: -0.1700

Data Index: 1597739 (Python Code)

Max Activating Token Index: 841

Text #15

Max Range: 4.4578. Min Range: -4.4578

Max Act: 3.9730. Min Act: -0.1696

Data Index: 1458260 (Python Code)

Max Activating Token Index: 755

Text #16

Max Range: 4.4578. Min Range: -4.4578

Max Act: 3.9641. Min Act: -0.1700

Data Index: 1424123 (Python Code)

Max Activating Token Index: 930

Text #17

Max Range: 4.4578. Min Range: -4.4578

Max Act: 3.9887. Min Act: -0.1700

Data Index: 1425556 (Python Code)

Max Activating Token Index: 784

Text #18

Max Range: 4.4578. Min Range: -4.4578

Max Act: 3.9662. Min Act: -0.1700

Data Index: 1429598 (Python Code)

Max Activating Token Index: 540

Text #19

Max Range: 4.4578. Min Range: -4.4578

Max Act: 3.9729. Min Act: -0.1696

Data Index: 481325 (C4 (Web Text))

Max Activating Token Index: 419