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

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

Neuron 1951 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: 5.8867. Min Range: -5.8867

Max Act: 5.8867. Min Act: -0.1700

Data Index: 1105986 (C4 (Web Text))

Max Activating Token Index: 890

Text #1

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.7969. Min Act: -0.1700

Data Index: 256490 (C4 (Web Text))

Max Activating Token Index: 995

Text #2

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.7422. Min Act: -0.1700

Data Index: 980539 (C4 (Web Text))

Max Activating Token Index: 664

Text #3

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.6992. Min Act: -0.1700

Data Index: 370903 (C4 (Web Text))

Max Activating Token Index: 205

Text #4

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.7070. Min Act: -0.1700

Data Index: 840312 (C4 (Web Text))

Max Activating Token Index: 889

Text #5

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.6562. Min Act: -0.1700

Data Index: 999632 (C4 (Web Text))

Max Activating Token Index: 739

Text #6

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.6601. Min Act: -0.1700

Data Index: 1159247 (C4 (Web Text))

Max Activating Token Index: 410

Text #7

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.5644. Min Act: -0.1700

Data Index: 556346 (C4 (Web Text))

Max Activating Token Index: 626

Text #8

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.4238. Min Act: -0.1700

Data Index: 862959 (C4 (Web Text))

Max Activating Token Index: 834

Text #9

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.4414. Min Act: -0.1700

Data Index: 375325 (C4 (Web Text))

Max Activating Token Index: 757

Text #10

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.5078. Min Act: -0.1700

Data Index: 1139601 (C4 (Web Text))

Max Activating Token Index: 129

Text #11

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.3945. Min Act: -0.1700

Data Index: 1015694 (C4 (Web Text))

Max Activating Token Index: 276

Text #12

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.3506. Min Act: -0.1700

Data Index: 726765 (C4 (Web Text))

Max Activating Token Index: 632

Text #13

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.3945. Min Act: -0.1700

Data Index: 909609 (C4 (Web Text))

Max Activating Token Index: 281

Text #14

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.4023. Min Act: -0.1700

Data Index: 65168 (C4 (Web Text))

Max Activating Token Index: 258

Text #15

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.3613. Min Act: -0.1700

Data Index: 1128371 (C4 (Web Text))

Max Activating Token Index: 904

Text #16

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.3594. Min Act: -0.1700

Data Index: 522235 (C4 (Web Text))

Max Activating Token Index: 922

Text #17

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.2241. Min Act: -0.1700

Data Index: 599959 (C4 (Web Text))

Max Activating Token Index: 64

Text #18

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.2148. Min Act: -0.1700

Data Index: 1041933 (C4 (Web Text))

Max Activating Token Index: 528

Text #19

Max Range: 5.8867. Min Range: -5.8867

Max Act: 5.1745. Min Act: -0.1700

Data Index: 189252 (C4 (Web Text))

Max Activating Token Index: 528