Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1800 in Layer 2
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.6673. Min Act: -0.1700
Data Index: 1537526 (Python Code)
Max Activating Token Index: 749
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.6361. Min Act: -0.1700
Data Index: 1241546 (C4 (Web Text))
Max Activating Token Index: 710
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.7298. Min Act: -0.1700
Data Index: 1560348 (Python Code)
Max Activating Token Index: 676
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.5736. Min Act: -0.1700
Data Index: 1623304 (Python Code)
Max Activating Token Index: 693
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.5423. Min Act: -0.1700
Data Index: 1291923 (C4 (Web Text))
Max Activating Token Index: 448
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.5111. Min Act: -0.1700
Data Index: 1564367 (Python Code)
Max Activating Token Index: 756
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4486. Min Act: -0.1700
Data Index: 807551 (C4 (Web Text))
Max Activating Token Index: 588
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4486. Min Act: -0.1700
Data Index: 1581113 (Python Code)
Max Activating Token Index: 774
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4486. Min Act: -0.1700
Data Index: 1614305 (Python Code)
Max Activating Token Index: 795
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4798. Min Act: -0.1700
Data Index: 1441386 (Python Code)
Max Activating Token Index: 921
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4486. Min Act: -0.1700
Data Index: 69923 (C4 (Web Text))
Max Activating Token Index: 226
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4798. Min Act: -0.1700
Data Index: 149515 (C4 (Web Text))
Max Activating Token Index: 687
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4798. Min Act: -0.1700
Data Index: 1389331 (Python Code)
Max Activating Token Index: 695
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4798. Min Act: -0.1700
Data Index: 1470805 (Python Code)
Max Activating Token Index: 797
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4798. Min Act: -0.1700
Data Index: 132331 (C4 (Web Text))
Max Activating Token Index: 291
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4798. Min Act: -0.1700
Data Index: 309954 (C4 (Web Text))
Max Activating Token Index: 294
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4486. Min Act: -0.1700
Data Index: 189251 (C4 (Web Text))
Max Activating Token Index: 613
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4486. Min Act: -0.1700
Data Index: 414592 (C4 (Web Text))
Max Activating Token Index: 817
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.3861. Min Act: -0.1700
Data Index: 1653583 (Python Code)
Max Activating Token Index: 661
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.7298. Min Range: -4.7298
Max Act: 4.4486. Min Act: -0.1700
Data Index: 1388908 (Python Code)
Max Activating Token Index: 653
Click toggle to see full text
Truncated
Full Text #19