Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1124 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.9852. Min Act: -0.1700

Data Index: 832647 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.9539. Min Act: -0.1700

Data Index: 141435 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.7976. Min Act: -0.1700

Data Index: 1392781 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.7037. Min Act: -0.1700

Data Index: 512245 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.6411. Min Act: -0.1700

Data Index: 872778 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.6098. Min Act: -0.1700

Data Index: 982185 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.6098. Min Act: -0.1700

Data Index: 1483065 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.6098. Min Act: -0.1700

Data Index: 390079 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.5785. Min Act: -0.1700

Data Index: 419980 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.5471. Min Act: -0.1700

Data Index: 227024 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.5158. Min Act: -0.1699

Data Index: 1574441 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.5158. Min Act: -0.1700

Data Index: 250881 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.5158. Min Act: -0.1700

Data Index: 1533309 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.4531. Min Act: -0.1700

Data Index: 1499947 (Python Code)

Max Activating Token Index: 791

Text #14

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.4218. Min Act: -0.1700

Data Index: 1479591 (Python Code)

Max Activating Token Index: 871

Text #15

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.3590. Min Act: -0.1700

Data Index: 705213 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.3590. Min Act: -0.1700

Data Index: 1080474 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.3590. Min Act: -0.1700

Data Index: 1071425 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.3276. Min Act: -0.1700

Data Index: 1323605 (C4 (Web Text))

Max Activating Token Index: 944

Text #19

Max Range: 3.9852. Min Range: -3.9852

Max Act: 3.3590. Min Act: -0.1700

Data Index: 1439768 (Python Code)

Max Activating Token Index: 1023