Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1997 in Layer 3
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.1841. Min Act: -0.1700
Data Index: 261814 (C4 (Web Text))
Max Activating Token Index: 592
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.1841. Min Act: -0.1700
Data Index: 23124 (C4 (Web Text))
Max Activating Token Index: 907
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.1528. Min Act: -0.1700
Data Index: 832810 (C4 (Web Text))
Max Activating Token Index: 958
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.1215. Min Act: -0.1700
Data Index: 570316 (C4 (Web Text))
Max Activating Token Index: 820
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.1215. Min Act: -0.1700
Data Index: 1242389 (C4 (Web Text))
Max Activating Token Index: 971
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.1528. Min Act: -0.1700
Data Index: 998424 (C4 (Web Text))
Max Activating Token Index: 290
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.0278. Min Act: -0.1700
Data Index: 389457 (C4 (Web Text))
Max Activating Token Index: 446
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.1841. Min Range: -4.1841
Max Act: 3.9965. Min Act: -0.1700
Data Index: 220531 (C4 (Web Text))
Max Activating Token Index: 920
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.0278. Min Act: -0.1700
Data Index: 365026 (C4 (Web Text))
Max Activating Token Index: 647
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.0278. Min Act: -0.1700
Data Index: 1008325 (C4 (Web Text))
Max Activating Token Index: 678
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.1841. Min Range: -4.1841
Max Act: 3.9965. Min Act: -0.1700
Data Index: 1125312 (C4 (Web Text))
Max Activating Token Index: 969
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.0278. Min Act: -0.1700
Data Index: 394849 (C4 (Web Text))
Max Activating Token Index: 401
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.0278. Min Act: -0.1700
Data Index: 81034 (C4 (Web Text))
Max Activating Token Index: 63
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.0278. Min Act: -0.1700
Data Index: 1319814 (C4 (Web Text))
Max Activating Token Index: 579
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.1841. Min Range: -4.1841
Max Act: 4.0590. Min Act: -0.1700
Data Index: 1277477 (C4 (Web Text))
Max Activating Token Index: 301
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.1841. Min Range: -4.1841
Max Act: 3.9965. Min Act: -0.1700
Data Index: 576857 (C4 (Web Text))
Max Activating Token Index: 462
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.1841. Min Range: -4.1841
Max Act: 3.9965. Min Act: -0.1700
Data Index: 1011979 (C4 (Web Text))
Max Activating Token Index: 299
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.1841. Min Range: -4.1841
Max Act: 3.9965. Min Act: -0.1700
Data Index: 773576 (C4 (Web Text))
Max Activating Token Index: 195
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.1841. Min Range: -4.1841
Max Act: 3.9652. Min Act: -0.1700
Data Index: 150673 (C4 (Web Text))
Max Activating Token Index: 368
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.1841. Min Range: -4.1841
Max Act: 3.9965. Min Act: -0.1700
Data Index: 307012 (C4 (Web Text))
Max Activating Token Index: 112
Click toggle to see full text
Truncated
Full Text #19