Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 5 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.1557. Min Act: -0.1700
Data Index: 464756 (C4 (Web Text))
Max Activating Token Index: 591
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.1715. Min Act: -0.1700
Data Index: 208873 (C4 (Web Text))
Max Activating Token Index: 694
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.0611. Min Act: -0.1700
Data Index: 212115 (C4 (Web Text))
Max Activating Token Index: 741
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.0295. Min Act: -0.1700
Data Index: 1274786 (C4 (Web Text))
Max Activating Token Index: 88
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9979. Min Act: -0.1700
Data Index: 612871 (C4 (Web Text))
Max Activating Token Index: 176
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.0295. Min Act: -0.1700
Data Index: 943530 (C4 (Web Text))
Max Activating Token Index: 868
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9821. Min Act: -0.1700
Data Index: 155257 (C4 (Web Text))
Max Activating Token Index: 439
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.0137. Min Act: -0.1700
Data Index: 668404 (C4 (Web Text))
Max Activating Token Index: 812
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9505. Min Act: -0.1700
Data Index: 61585 (C4 (Web Text))
Max Activating Token Index: 711
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.0295. Min Act: -0.1700
Data Index: 1278496 (C4 (Web Text))
Max Activating Token Index: 326
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.0453. Min Act: -0.1700
Data Index: 961763 (C4 (Web Text))
Max Activating Token Index: 912
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9346. Min Act: -0.1700
Data Index: 483111 (C4 (Web Text))
Max Activating Token Index: 200
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9979. Min Act: -0.1700
Data Index: 771186 (C4 (Web Text))
Max Activating Token Index: 145
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9505. Min Act: -0.1700
Data Index: 504321 (C4 (Web Text))
Max Activating Token Index: 719
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9979. Min Act: -0.1700
Data Index: 375310 (C4 (Web Text))
Max Activating Token Index: 272
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 3.1715. Min Range: -3.1715
Max Act: 3.0137. Min Act: -0.1700
Data Index: 1036748 (C4 (Web Text))
Max Activating Token Index: 307
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9821. Min Act: -0.1700
Data Index: 325042 (C4 (Web Text))
Max Activating Token Index: 458
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9505. Min Act: -0.1700
Data Index: 152515 (C4 (Web Text))
Max Activating Token Index: 417
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9821. Min Act: -0.1700
Data Index: 333014 (C4 (Web Text))
Max Activating Token Index: 889
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 3.1715. Min Range: -3.1715
Max Act: 2.9346. Min Act: -0.1700
Data Index: 994976 (C4 (Web Text))
Max Activating Token Index: 381
Click toggle to see full text
Truncated
Full Text #19