Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 55 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.8938. Min Act: -0.1700
Data Index: 843035 (C4 (Web Text))
Max Activating Token Index: 649
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.7376. Min Act: -0.1700
Data Index: 645038 (C4 (Web Text))
Max Activating Token Index: 860
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.7063. Min Act: -0.1700
Data Index: 423013 (C4 (Web Text))
Max Activating Token Index: 258
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.7063. Min Act: -0.1700
Data Index: 119790 (C4 (Web Text))
Max Activating Token Index: 351
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.7063. Min Act: -0.1700
Data Index: 938794 (C4 (Web Text))
Max Activating Token Index: 591
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.6751. Min Act: -0.1700
Data Index: 169516 (C4 (Web Text))
Max Activating Token Index: 765
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.6751. Min Act: -0.1700
Data Index: 807775 (C4 (Web Text))
Max Activating Token Index: 454
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.6751. Min Act: -0.1700
Data Index: 257505 (C4 (Web Text))
Max Activating Token Index: 1014
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.6438. Min Act: -0.1700
Data Index: 465914 (C4 (Web Text))
Max Activating Token Index: 463
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.6438. Min Act: -0.1700
Data Index: 409793 (C4 (Web Text))
Max Activating Token Index: 519
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.6438. Min Act: -0.1700
Data Index: 1233502 (C4 (Web Text))
Max Activating Token Index: 801
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5813. Min Act: -0.1700
Data Index: 501692 (C4 (Web Text))
Max Activating Token Index: 557
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5813. Min Act: -0.1700
Data Index: 1209055 (C4 (Web Text))
Max Activating Token Index: 501
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5501. Min Act: -0.1700
Data Index: 1296413 (C4 (Web Text))
Max Activating Token Index: 50
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5501. Min Act: -0.1700
Data Index: 602983 (C4 (Web Text))
Max Activating Token Index: 942
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5813. Min Act: -0.1700
Data Index: 681269 (C4 (Web Text))
Max Activating Token Index: 589
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5501. Min Act: -0.1700
Data Index: 582946 (C4 (Web Text))
Max Activating Token Index: 513
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5813. Min Act: -0.1700
Data Index: 1220783 (C4 (Web Text))
Max Activating Token Index: 694
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5188. Min Act: -0.1700
Data Index: 644050 (C4 (Web Text))
Max Activating Token Index: 926
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.8938. Min Range: -4.8938
Max Act: 4.5501. Min Act: -0.1700
Data Index: 504220 (C4 (Web Text))
Max Activating Token Index: 335
Click toggle to see full text
Truncated
Full Text #19