Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 295 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.1961. Min Act: -0.1700

Data Index: 648839 (C4 (Web Text))

Max Activating Token Index: 152

Text #1

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.1492. Min Act: -0.1700

Data Index: 366933 (C4 (Web Text))

Max Activating Token Index: 455

Text #2

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.1648. Min Act: -0.1700

Data Index: 1124327 (C4 (Web Text))

Max Activating Token Index: 177

Text #3

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.0867. Min Act: -0.1700

Data Index: 933775 (C4 (Web Text))

Max Activating Token Index: 235

Text #4

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.1023. Min Act: -0.1700

Data Index: 759665 (C4 (Web Text))

Max Activating Token Index: 47

Text #5

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.1023. Min Act: -0.1700

Data Index: 606220 (C4 (Web Text))

Max Activating Token Index: 68

Text #6

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.0555. Min Act: -0.1700

Data Index: 1242540 (C4 (Web Text))

Max Activating Token Index: 72

Text #7

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.0398. Min Act: -0.1700

Data Index: 1188996 (C4 (Web Text))

Max Activating Token Index: 92

Text #8

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.0711. Min Act: -0.1700

Data Index: 656718 (C4 (Web Text))

Max Activating Token Index: 234

Text #9

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.0242. Min Act: -0.1700

Data Index: 69056 (C4 (Web Text))

Max Activating Token Index: 194

Text #10

Max Range: 5.1961. Min Range: -5.1961

Max Act: 5.0242. Min Act: -0.1700

Data Index: 296997 (C4 (Web Text))

Max Activating Token Index: 570

Text #11

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9930. Min Act: -0.1700

Data Index: 1087309 (C4 (Web Text))

Max Activating Token Index: 54

Text #12

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9617. Min Act: -0.1700

Data Index: 590124 (C4 (Web Text))

Max Activating Token Index: 601

Text #13

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9773. Min Act: -0.1700

Data Index: 1084678 (C4 (Web Text))

Max Activating Token Index: 107

Text #14

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9461. Min Act: -0.1700

Data Index: 1109276 (C4 (Web Text))

Max Activating Token Index: 426

Text #15

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9305. Min Act: -0.1700

Data Index: 621288 (C4 (Web Text))

Max Activating Token Index: 33

Text #16

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9617. Min Act: -0.1700

Data Index: 1313557 (C4 (Web Text))

Max Activating Token Index: 147

Text #17

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9461. Min Act: -0.1700

Data Index: 120222 (C4 (Web Text))

Max Activating Token Index: 289

Text #18

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9617. Min Act: -0.1700

Data Index: 849211 (C4 (Web Text))

Max Activating Token Index: 350

Text #19

Max Range: 5.1961. Min Range: -5.1961

Max Act: 4.9148. Min Act: -0.1700

Data Index: 942897 (C4 (Web Text))

Max Activating Token Index: 306