Model: SoLU Model: 8 Layers, 4096 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 2018 in Layer 4
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')
Text #0
Max Range: 2.3817. Min Range: -2.3817
Max Act: 2.3817. Min Act: -0.0001
Data Index: 146668 (C4 (Web Text))
Max Activating Token Index: 686
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.3817. Min Range: -2.3817
Max Act: 2.1538. Min Act: -0.0001
Data Index: 226133 (C4 (Web Text))
Max Activating Token Index: 817
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.3817. Min Range: -2.3817
Max Act: 2.0208. Min Act: -0.0001
Data Index: 225922 (C4 (Web Text))
Max Activating Token Index: 818
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.3817. Min Range: -2.3817
Max Act: 2.0281. Min Act: -0.0001
Data Index: 1053947 (C4 (Web Text))
Max Activating Token Index: 922
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.3817. Min Range: -2.3817
Max Act: 2.0887. Min Act: -0.0001
Data Index: 1273754 (C4 (Web Text))
Max Activating Token Index: 158
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.8042. Min Act: -0.0001
Data Index: 405517 (C4 (Web Text))
Max Activating Token Index: 889
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.8945. Min Act: -0.0001
Data Index: 1208470 (C4 (Web Text))
Max Activating Token Index: 510
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.8608. Min Act: -0.0001
Data Index: 979498 (C4 (Web Text))
Max Activating Token Index: 828
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.7155. Min Act: -0.0001
Data Index: 549911 (C4 (Web Text))
Max Activating Token Index: 627
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.7329. Min Act: -0.0001
Data Index: 290835 (C4 (Web Text))
Max Activating Token Index: 175
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.7233. Min Act: -0.0001
Data Index: 26743 (C4 (Web Text))
Max Activating Token Index: 533
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.8501. Min Act: -0.0001
Data Index: 261174 (C4 (Web Text))
Max Activating Token Index: 951
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.7710. Min Act: -0.0001
Data Index: 55670 (C4 (Web Text))
Max Activating Token Index: 585
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.7983. Min Act: -0.0001
Data Index: 946375 (C4 (Web Text))
Max Activating Token Index: 738
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.7399. Min Act: -0.0001
Data Index: 126425 (C4 (Web Text))
Max Activating Token Index: 855
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.6369. Min Act: -0.0001
Data Index: 1135020 (C4 (Web Text))
Max Activating Token Index: 531
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.6227. Min Act: -0.0001
Data Index: 324530 (C4 (Web Text))
Max Activating Token Index: 865
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.7061. Min Act: -0.0001
Data Index: 813645 (C4 (Web Text))
Max Activating Token Index: 933
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.5798. Min Act: -0.0001
Data Index: 112353 (C4 (Web Text))
Max Activating Token Index: 504
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.3817. Min Range: -2.3817
Max Act: 1.3792. Min Act: -0.0001
Data Index: 833819 (C4 (Web Text))
Max Activating Token Index: 615
Click toggle to see full text
Truncated
Full Text #19