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

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

Neuron 1712 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: 7.8030. Min Range: -7.8030

Max Act: 7.8030. Min Act: -0.1700

Data Index: 1577804 (Python Code)

Max Activating Token Index: 399

Text #1

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.7432. Min Act: -0.1700

Data Index: 1515251 (Python Code)

Max Activating Token Index: 774

Text #2

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.7574. Min Act: -0.1700

Data Index: 1649599 (Python Code)

Max Activating Token Index: 6

Text #3

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.7307. Min Act: -0.1700

Data Index: 1496901 (Python Code)

Max Activating Token Index: 171

Text #4

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.7135. Min Act: -0.1700

Data Index: 1494864 (Python Code)

Max Activating Token Index: 336

Text #5

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.4420. Min Act: -0.1700

Data Index: 1591526 (Python Code)

Max Activating Token Index: 112

Text #6

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.3075. Min Act: -0.1700

Data Index: 1566606 (Python Code)

Max Activating Token Index: 392

Text #7

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.3580. Min Act: -0.1700

Data Index: 1631409 (Python Code)

Max Activating Token Index: 991

Text #8

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.3498. Min Act: -0.1700

Data Index: 1416727 (Python Code)

Max Activating Token Index: 274

Text #9

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.3154. Min Act: -0.1700

Data Index: 1587179 (Python Code)

Max Activating Token Index: 305

Text #10

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2373. Min Act: -0.1700

Data Index: 1488072 (Python Code)

Max Activating Token Index: 624

Text #11

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2377. Min Act: -0.1700

Data Index: 1624295 (Python Code)

Max Activating Token Index: 477

Text #12

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2438. Min Act: -0.1700

Data Index: 1595008 (Python Code)

Max Activating Token Index: 250

Text #13

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2232. Min Act: -0.1700

Data Index: 1604378 (Python Code)

Max Activating Token Index: 2

Text #14

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2144. Min Act: -0.1700

Data Index: 1440995 (Python Code)

Max Activating Token Index: 154

Text #15

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.1625. Min Act: -0.1700

Data Index: 1606025 (Python Code)

Max Activating Token Index: 2

Text #16

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2100. Min Act: -0.1700

Data Index: 1430269 (Python Code)

Max Activating Token Index: 167

Text #17

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2299. Min Act: -0.1700

Data Index: 1590236 (Python Code)

Max Activating Token Index: 524

Text #18

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2292. Min Act: -0.1700

Data Index: 1495261 (Python Code)

Max Activating Token Index: 7

Text #19

Max Range: 7.8030. Min Range: -7.8030

Max Act: 7.2318. Min Act: -0.1700

Data Index: 1567365 (Python Code)

Max Activating Token Index: 801