Model: SoLU Model: 10 Layers, 5120 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 4931 in Layer 4
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')
Text #0
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.3851. Min Act: -0.0001
Data Index: 872754 (C4 (Web Text))
Max Activating Token Index: 40
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.3705. Min Act: -0.0001
Data Index: 1197495 (C4 (Web Text))
Max Activating Token Index: 143
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.2748. Min Act: -0.0001
Data Index: 108002 (C4 (Web Text))
Max Activating Token Index: 612
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.1581. Min Act: -0.0001
Data Index: 54946 (C4 (Web Text))
Max Activating Token Index: 31
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.2166. Min Act: -0.0001
Data Index: 255029 (C4 (Web Text))
Max Activating Token Index: 372
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.1442. Min Act: -0.0001
Data Index: 229760 (C4 (Web Text))
Max Activating Token Index: 338
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.2105. Min Act: -0.0001
Data Index: 711789 (C4 (Web Text))
Max Activating Token Index: 229
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.1679. Min Act: -0.0001
Data Index: 666866 (C4 (Web Text))
Max Activating Token Index: 505
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.2735. Min Act: -0.0001
Data Index: 973977 (C4 (Web Text))
Max Activating Token Index: 548
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.2140. Min Act: -0.0001
Data Index: 588104 (C4 (Web Text))
Max Activating Token Index: 705
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.0892. Min Act: -0.0001
Data Index: 191690 (C4 (Web Text))
Max Activating Token Index: 792
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.0988. Min Act: -0.0001
Data Index: 129724 (C4 (Web Text))
Max Activating Token Index: 50
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.1604. Min Act: -0.0001
Data Index: 214331 (C4 (Web Text))
Max Activating Token Index: 166
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.1493. Min Act: -0.0001
Data Index: 537977 (C4 (Web Text))
Max Activating Token Index: 129
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.0648. Min Act: -0.0001
Data Index: 1603370 (Python Code)
Max Activating Token Index: 759
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.0940. Min Act: -0.0001
Data Index: 236124 (C4 (Web Text))
Max Activating Token Index: 914
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.1386. Min Act: -0.0001
Data Index: 892722 (C4 (Web Text))
Max Activating Token Index: 97
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 1.3851. Min Range: -1.3851
Max Act: 0.9963. Min Act: -0.0001
Data Index: 463542 (C4 (Web Text))
Max Activating Token Index: 599
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 1.3851. Min Range: -1.3851
Max Act: 0.9971. Min Act: -0.0001
Data Index: 1094471 (C4 (Web Text))
Max Activating Token Index: 730
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 1.3851. Min Range: -1.3851
Max Act: 1.0083. Min Act: -0.0001
Data Index: 1395600 (Python Code)
Max Activating Token Index: 25
Click toggle to see full text
Truncated
Full Text #19