Model: SoLU Model: 8 Layers, 4096 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1600 in Layer 6
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')
Text #0
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.4918. Min Act: -0.0001
Data Index: 763563 (C4 (Web Text))
Max Activating Token Index: 389
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.3258. Min Act: -0.0001
Data Index: 517637 (C4 (Web Text))
Max Activating Token Index: 634
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.3420. Min Act: -0.0001
Data Index: 455942 (C4 (Web Text))
Max Activating Token Index: 608
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.2489. Min Act: -0.0001
Data Index: 1387655 (Python Code)
Max Activating Token Index: 866
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.2409. Min Act: -0.0001
Data Index: 143567 (C4 (Web Text))
Max Activating Token Index: 545
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.2682. Min Act: -0.0001
Data Index: 1022975 (C4 (Web Text))
Max Activating Token Index: 843
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.2211. Min Act: -0.0001
Data Index: 1420021 (Python Code)
Max Activating Token Index: 897
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.2813. Min Act: -0.0001
Data Index: 605745 (C4 (Web Text))
Max Activating Token Index: 792
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.1567. Min Act: -0.0001
Data Index: 425163 (C4 (Web Text))
Max Activating Token Index: 722
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.3284. Min Act: -0.0001
Data Index: 1402872 (Python Code)
Max Activating Token Index: 690
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.2159. Min Act: -0.0001
Data Index: 1506008 (Python Code)
Max Activating Token Index: 893
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.2020. Min Act: -0.0001
Data Index: 490844 (C4 (Web Text))
Max Activating Token Index: 137
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.1792. Min Act: -0.0001
Data Index: 1364490 (Python Code)
Max Activating Token Index: 812
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.2034. Min Act: -0.0001
Data Index: 1484157 (Python Code)
Max Activating Token Index: 946
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.1016. Min Act: -0.0001
Data Index: 1385193 (Python Code)
Max Activating Token Index: 947
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.0894. Min Act: -0.0001
Data Index: 426649 (C4 (Web Text))
Max Activating Token Index: 482
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.0007. Min Act: -0.0001
Data Index: 1418320 (Python Code)
Max Activating Token Index: 85
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.1149. Min Act: -0.0001
Data Index: 1465502 (Python Code)
Max Activating Token Index: 618
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 1.4918. Min Range: -1.4918
Max Act: 1.0354. Min Act: -0.0001
Data Index: 1309088 (C4 (Web Text))
Max Activating Token Index: 206
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 1.4918. Min Range: -1.4918
Max Act: 0.9806. Min Act: -0.0001
Data Index: 1526656 (Python Code)
Max Activating Token Index: 923
Click toggle to see full text
Truncated
Full Text #19