Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 159 in Layer 0
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.8448. Min Act: -0.1700
Data Index: 487408 (C4 (Web Text))
Max Activating Token Index: 286
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.8289. Min Act: -0.1700
Data Index: 1027315 (C4 (Web Text))
Max Activating Token Index: 872
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.8130. Min Act: -0.1700
Data Index: 1354895 (C4 (Web Text))
Max Activating Token Index: 610
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.8130. Min Act: -0.1700
Data Index: 881343 (C4 (Web Text))
Max Activating Token Index: 1002
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7971. Min Act: -0.1700
Data Index: 674595 (C4 (Web Text))
Max Activating Token Index: 464
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7492. Min Act: -0.1700
Data Index: 127516 (C4 (Web Text))
Max Activating Token Index: 49
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7652. Min Act: -0.1700
Data Index: 911077 (C4 (Web Text))
Max Activating Token Index: 515
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7652. Min Act: -0.1700
Data Index: 850133 (C4 (Web Text))
Max Activating Token Index: 11
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7492. Min Act: -0.1700
Data Index: 126469 (C4 (Web Text))
Max Activating Token Index: 96
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7492. Min Act: -0.1700
Data Index: 520435 (C4 (Web Text))
Max Activating Token Index: 815
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7652. Min Act: -0.1700
Data Index: 1166935 (C4 (Web Text))
Max Activating Token Index: 709
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7333. Min Act: -0.1700
Data Index: 146036 (C4 (Web Text))
Max Activating Token Index: 91
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7492. Min Act: -0.1700
Data Index: 1077641 (C4 (Web Text))
Max Activating Token Index: 751
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7652. Min Act: -0.1700
Data Index: 655673 (C4 (Web Text))
Max Activating Token Index: 97
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7173. Min Act: -0.1700
Data Index: 364866 (C4 (Web Text))
Max Activating Token Index: 4
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7333. Min Act: -0.1700
Data Index: 1047178 (C4 (Web Text))
Max Activating Token Index: 740
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7333. Min Act: -0.1700
Data Index: 1039114 (C4 (Web Text))
Max Activating Token Index: 611
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7333. Min Act: -0.1700
Data Index: 945206 (C4 (Web Text))
Max Activating Token Index: 148
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7173. Min Act: -0.1700
Data Index: 155468 (C4 (Web Text))
Max Activating Token Index: 458
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.8448. Min Range: -2.8448
Max Act: 2.7173. Min Act: -0.1700
Data Index: 627829 (C4 (Web Text))
Max Activating Token Index: 50
Click toggle to see full text
Truncated
Full Text #19