Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1972 in Layer 3
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.8083. Min Act: -0.1700
Data Index: 495662 (C4 (Web Text))
Max Activating Token Index: 24
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.6676. Min Act: -0.1700
Data Index: 340080 (C4 (Web Text))
Max Activating Token Index: 302
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.6833. Min Act: -0.1700
Data Index: 604058 (C4 (Web Text))
Max Activating Token Index: 797
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.6051. Min Act: -0.1700
Data Index: 193214 (C4 (Web Text))
Max Activating Token Index: 641
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.6208. Min Act: -0.1700
Data Index: 1009800 (C4 (Web Text))
Max Activating Token Index: 487
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.6208. Min Act: -0.1700
Data Index: 603886 (C4 (Web Text))
Max Activating Token Index: 768
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.6208. Min Act: -0.1700
Data Index: 314961 (C4 (Web Text))
Max Activating Token Index: 280
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5583. Min Act: -0.1700
Data Index: 821869 (C4 (Web Text))
Max Activating Token Index: 376
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.6051. Min Act: -0.1700
Data Index: 157400 (C4 (Web Text))
Max Activating Token Index: 997
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.6051. Min Act: -0.1700
Data Index: 866138 (C4 (Web Text))
Max Activating Token Index: 829
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5583. Min Act: -0.1700
Data Index: 907793 (C4 (Web Text))
Max Activating Token Index: 998
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5739. Min Act: -0.1700
Data Index: 556813 (C4 (Web Text))
Max Activating Token Index: 28
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5426. Min Act: -0.1700
Data Index: 728896 (C4 (Web Text))
Max Activating Token Index: 1020
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5270. Min Act: -0.1700
Data Index: 154668 (C4 (Web Text))
Max Activating Token Index: 783
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5426. Min Act: -0.1700
Data Index: 626222 (C4 (Web Text))
Max Activating Token Index: 497
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5270. Min Act: -0.1700
Data Index: 108180 (C4 (Web Text))
Max Activating Token Index: 13
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5583. Min Act: -0.1700
Data Index: 919818 (C4 (Web Text))
Max Activating Token Index: 385
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5583. Min Act: -0.1700
Data Index: 1259212 (C4 (Web Text))
Max Activating Token Index: 905
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5114. Min Act: -0.1700
Data Index: 953515 (C4 (Web Text))
Max Activating Token Index: 464
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.8083. Min Range: -4.8083
Max Act: 4.5583. Min Act: -0.1700
Data Index: 187796 (C4 (Web Text))
Max Activating Token Index: 889
Click toggle to see full text
Truncated
Full Text #19