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

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

Neuron 1782 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.8904. Min Act: -0.1700

Data Index: 669826 (C4 (Web Text))

Max Activating Token Index: 634

Text #1

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.8591. Min Act: -0.1700

Data Index: 118710 (C4 (Web Text))

Max Activating Token Index: 803

Text #2

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.7029. Min Act: -0.1700

Data Index: 891999 (C4 (Web Text))

Max Activating Token Index: 717

Text #3

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.7654. Min Act: -0.1700

Data Index: 65008 (C4 (Web Text))

Max Activating Token Index: 85

Text #4

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.6716. Min Act: -0.1700

Data Index: 982242 (C4 (Web Text))

Max Activating Token Index: 300

Text #5

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.6716. Min Act: -0.1700

Data Index: 833123 (C4 (Web Text))

Max Activating Token Index: 411

Text #6

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.6091. Min Act: -0.1700

Data Index: 531489 (C4 (Web Text))

Max Activating Token Index: 364

Text #7

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.5466. Min Act: -0.1700

Data Index: 90634 (C4 (Web Text))

Max Activating Token Index: 652

Text #8

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.5779. Min Act: -0.1700

Data Index: 1259280 (C4 (Web Text))

Max Activating Token Index: 217

Text #9

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.5466. Min Act: -0.1700

Data Index: 627375 (C4 (Web Text))

Max Activating Token Index: 967

Text #10

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.4685. Min Act: -0.1700

Data Index: 60633 (C4 (Web Text))

Max Activating Token Index: 170

Text #11

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.4216. Min Act: -0.1700

Data Index: 71630 (C4 (Web Text))

Max Activating Token Index: 950

Text #12

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.3904. Min Act: -0.1700

Data Index: 729907 (C4 (Web Text))

Max Activating Token Index: 625

Text #13

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.4060. Min Act: -0.1700

Data Index: 47924 (C4 (Web Text))

Max Activating Token Index: 705

Text #14

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.3279. Min Act: -0.1700

Data Index: 133612 (C4 (Web Text))

Max Activating Token Index: 500

Text #15

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.4060. Min Act: -0.1700

Data Index: 132086 (C4 (Web Text))

Max Activating Token Index: 672

Text #16

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.3904. Min Act: -0.1700

Data Index: 256253 (C4 (Web Text))

Max Activating Token Index: 277

Text #17

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.3591. Min Act: -0.1700

Data Index: 94158 (C4 (Web Text))

Max Activating Token Index: 427

Text #18

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.3904. Min Act: -0.1700

Data Index: 972498 (C4 (Web Text))

Max Activating Token Index: 959

Text #19

Max Range: 5.8904. Min Range: -5.8904

Max Act: 5.3435. Min Act: -0.1700

Data Index: 335955 (C4 (Web Text))

Max Activating Token Index: 513