Model: SoLU Model: 10 Layers, 5120 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 2750 in Layer 4
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')
Text #0
Max Range: 2.4898. Min Range: -2.4898
Max Act: 2.3821. Min Act: -0.0001
Data Index: 612942 (C4 (Web Text))
Max Activating Token Index: 125
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.4898. Min Range: -2.4898
Max Act: 2.3267. Min Act: -0.0001
Data Index: 27110 (C4 (Web Text))
Max Activating Token Index: 841
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.4898. Min Range: -2.4898
Max Act: 2.4303. Min Act: -0.0001
Data Index: 1170983 (C4 (Web Text))
Max Activating Token Index: 414
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.4898. Min Range: -2.4898
Max Act: 2.4898. Min Act: -0.0001
Data Index: 1229753 (C4 (Web Text))
Max Activating Token Index: 737
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.4898. Min Range: -2.4898
Max Act: 2.0633. Min Act: -0.0001
Data Index: 9652 (C4 (Web Text))
Max Activating Token Index: 832
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.4898. Min Range: -2.4898
Max Act: 2.4161. Min Act: -0.0001
Data Index: 26200 (C4 (Web Text))
Max Activating Token Index: 921
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.8097. Min Act: -0.0001
Data Index: 18140 (C4 (Web Text))
Max Activating Token Index: 87
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.4898. Min Range: -2.4898
Max Act: 2.0499. Min Act: -0.0001
Data Index: 1521049 (Python Code)
Max Activating Token Index: 19
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.7815. Min Act: -0.0001
Data Index: 1374273 (Python Code)
Max Activating Token Index: 32
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.6330. Min Act: -0.0001
Data Index: 1292227 (C4 (Web Text))
Max Activating Token Index: 746
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.6763. Min Act: -0.0001
Data Index: 197592 (C4 (Web Text))
Max Activating Token Index: 802
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.6735. Min Act: -0.0001
Data Index: 773709 (C4 (Web Text))
Max Activating Token Index: 104
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.8800. Min Act: -0.0001
Data Index: 546729 (C4 (Web Text))
Max Activating Token Index: 801
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.6526. Min Act: -0.0001
Data Index: 1065330 (C4 (Web Text))
Max Activating Token Index: 73
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.6088. Min Act: -0.0001
Data Index: 1193240 (C4 (Web Text))
Max Activating Token Index: 802
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.7539. Min Act: -0.0001
Data Index: 1580453 (Python Code)
Max Activating Token Index: 443
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.6646. Min Act: -0.0001
Data Index: 496507 (C4 (Web Text))
Max Activating Token Index: 73
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.7306. Min Act: -0.0001
Data Index: 1577629 (Python Code)
Max Activating Token Index: 935
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.6111. Min Act: -0.0001
Data Index: 1503742 (Python Code)
Max Activating Token Index: 1004
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.4898. Min Range: -2.4898
Max Act: 1.5901. Min Act: -0.0001
Data Index: 1476751 (Python Code)
Max Activating Token Index: 765
Click toggle to see full text
Truncated
Full Text #19