Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 978 in Layer 0
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.6591. Min Act: -0.1700
Data Index: 741861 (C4 (Web Text))
Max Activating Token Index: 715
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.6270. Min Act: -0.1700
Data Index: 33873 (C4 (Web Text))
Max Activating Token Index: 88
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5787. Min Act: -0.1700
Data Index: 1066933 (C4 (Web Text))
Max Activating Token Index: 57
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5626. Min Act: -0.1700
Data Index: 734719 (C4 (Web Text))
Max Activating Token Index: 10
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5626. Min Act: -0.1700
Data Index: 676618 (C4 (Web Text))
Max Activating Token Index: 450
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5626. Min Act: -0.1700
Data Index: 770702 (C4 (Web Text))
Max Activating Token Index: 774
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5303. Min Act: -0.1700
Data Index: 681924 (C4 (Web Text))
Max Activating Token Index: 326
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5303. Min Act: -0.1700
Data Index: 245127 (C4 (Web Text))
Max Activating Token Index: 1018
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.4980. Min Act: -0.1700
Data Index: 1527522 (Python Code)
Max Activating Token Index: 962
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5141. Min Act: -0.1700
Data Index: 67803 (C4 (Web Text))
Max Activating Token Index: 142
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5141. Min Act: -0.1700
Data Index: 405977 (C4 (Web Text))
Max Activating Token Index: 14
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5141. Min Act: -0.1700
Data Index: 1028536 (C4 (Web Text))
Max Activating Token Index: 873
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.4980. Min Act: -0.1700
Data Index: 611729 (C4 (Web Text))
Max Activating Token Index: 5
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.4980. Min Act: -0.1700
Data Index: 1193894 (C4 (Web Text))
Max Activating Token Index: 136
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.4980. Min Act: -0.1700
Data Index: 373870 (C4 (Web Text))
Max Activating Token Index: 374
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.4980. Min Act: -0.1700
Data Index: 954803 (C4 (Web Text))
Max Activating Token Index: 327
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.5141. Min Act: -0.1700
Data Index: 230916 (C4 (Web Text))
Max Activating Token Index: 78
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.4980. Min Act: -0.1700
Data Index: 1115360 (C4 (Web Text))
Max Activating Token Index: 513
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.4818. Min Act: -0.1700
Data Index: 1455564 (Python Code)
Max Activating Token Index: 582
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.6591. Min Range: -2.6591
Max Act: 2.4980. Min Act: -0.1700
Data Index: 1575249 (Python Code)
Max Activating Token Index: 636
Click toggle to see full text
Truncated
Full Text #19