Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1805 in Layer 2
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 6.4703. Min Range: -6.4703
Max Act: 6.4703. Min Act: -0.1700
Data Index: 1059383 (C4 (Web Text))
Max Activating Token Index: 299
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 6.4703. Min Range: -6.4703
Max Act: 6.4078. Min Act: -0.1700
Data Index: 1096125 (C4 (Web Text))
Max Activating Token Index: 843
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 6.4703. Min Range: -6.4703
Max Act: 6.4546. Min Act: -0.1700
Data Index: 364228 (C4 (Web Text))
Max Activating Token Index: 420
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 6.4703. Min Range: -6.4703
Max Act: 6.4390. Min Act: -0.1700
Data Index: 1132936 (C4 (Web Text))
Max Activating Token Index: 597
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 6.4703. Min Range: -6.4703
Max Act: 6.2359. Min Act: -0.1700
Data Index: 1216414 (C4 (Web Text))
Max Activating Token Index: 216
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 6.4703. Min Range: -6.4703
Max Act: 6.0640. Min Act: -0.1700
Data Index: 1584984 (Python Code)
Max Activating Token Index: 923
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 6.4703. Min Range: -6.4703
Max Act: 6.0015. Min Act: -0.1700
Data Index: 1648658 (Python Code)
Max Activating Token Index: 912
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.8218. Min Act: -0.1700
Data Index: 387351 (C4 (Web Text))
Max Activating Token Index: 737
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.8531. Min Act: -0.1700
Data Index: 483066 (C4 (Web Text))
Max Activating Token Index: 955
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.7359. Min Act: -0.1700
Data Index: 1061345 (C4 (Web Text))
Max Activating Token Index: 189
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.6421. Min Act: -0.1700
Data Index: 725373 (C4 (Web Text))
Max Activating Token Index: 460
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.6968. Min Act: -0.1700
Data Index: 1398623 (Python Code)
Max Activating Token Index: 1016
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.7359. Min Act: -0.1700
Data Index: 1476395 (Python Code)
Max Activating Token Index: 456
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.7281. Min Act: -0.1700
Data Index: 744621 (C4 (Web Text))
Max Activating Token Index: 673
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.7125. Min Act: -0.1700
Data Index: 1513375 (Python Code)
Max Activating Token Index: 869
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.7515. Min Act: -0.1700
Data Index: 152612 (C4 (Web Text))
Max Activating Token Index: 930
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.6812. Min Act: -0.1700
Data Index: 1539217 (Python Code)
Max Activating Token Index: 818
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.6734. Min Act: -0.1700
Data Index: 878787 (C4 (Web Text))
Max Activating Token Index: 757
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.6500. Min Act: -0.1700
Data Index: 463869 (C4 (Web Text))
Max Activating Token Index: 692
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 6.4703. Min Range: -6.4703
Max Act: 5.7359. Min Act: -0.1700
Data Index: 1475039 (Python Code)
Max Activating Token Index: 864
Click toggle to see full text
Truncated
Full Text #19