Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 610 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.8587. Min Act: -0.1700

Data Index: 826501 (C4 (Web Text))

Max Activating Token Index: 243

Text #1

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.8275. Min Act: -0.1700

Data Index: 823098 (C4 (Web Text))

Max Activating Token Index: 230

Text #2

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.7650. Min Act: -0.1700

Data Index: 314316 (C4 (Web Text))

Max Activating Token Index: 414

Text #3

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.5462. Min Act: -0.1700

Data Index: 805417 (C4 (Web Text))

Max Activating Token Index: 356

Text #4

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.5150. Min Act: -0.1700

Data Index: 1291678 (C4 (Web Text))

Max Activating Token Index: 423

Text #5

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.4837. Min Act: -0.1700

Data Index: 1433502 (Python Code)

Max Activating Token Index: 155

Text #6

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.4837. Min Act: -0.1700

Data Index: 1054067 (C4 (Web Text))

Max Activating Token Index: 708

Text #7

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.4525. Min Act: -0.1700

Data Index: 161395 (C4 (Web Text))

Max Activating Token Index: 327

Text #8

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.4837. Min Act: -0.1700

Data Index: 334651 (C4 (Web Text))

Max Activating Token Index: 156

Text #9

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.3900. Min Act: -0.1700

Data Index: 1448382 (Python Code)

Max Activating Token Index: 311

Text #10

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.3900. Min Act: -0.1700

Data Index: 1565224 (Python Code)

Max Activating Token Index: 820

Text #11

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.4525. Min Act: -0.1700

Data Index: 1575079 (Python Code)

Max Activating Token Index: 336

Text #12

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.3900. Min Act: -0.1700

Data Index: 1655834 (Python Code)

Max Activating Token Index: 633

Text #13

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.3587. Min Act: -0.1700

Data Index: 732056 (C4 (Web Text))

Max Activating Token Index: 270

Text #14

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.2962. Min Act: -0.1700

Data Index: 1632329 (Python Code)

Max Activating Token Index: 513

Text #15

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.2962. Min Act: -0.1700

Data Index: 1631777 (Python Code)

Max Activating Token Index: 73

Text #16

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.3275. Min Act: -0.1700

Data Index: 1434713 (Python Code)

Max Activating Token Index: 110

Text #17

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.2962. Min Act: -0.1700

Data Index: 1649866 (Python Code)

Max Activating Token Index: 742

Text #18

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.2962. Min Act: -0.1685

Data Index: 1485669 (Python Code)

Max Activating Token Index: 1015

Text #19

Max Range: 5.8587. Min Range: -5.8587

Max Act: 5.2650. Min Act: -0.1699

Data Index: 1502144 (Python Code)

Max Activating Token Index: 263