Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1455 in Layer 2
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.4844. Min Act: -0.1700
Data Index: 731947 (C4 (Web Text))
Max Activating Token Index: 469
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.3907. Min Act: -0.1700
Data Index: 1370710 (Python Code)
Max Activating Token Index: 332
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.3750. Min Act: -0.1700
Data Index: 870595 (C4 (Web Text))
Max Activating Token Index: 484
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2813. Min Act: -0.1700
Data Index: 430303 (C4 (Web Text))
Max Activating Token Index: 948
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.3125. Min Act: -0.1700
Data Index: 529874 (C4 (Web Text))
Max Activating Token Index: 1008
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2500. Min Act: -0.1700
Data Index: 1644879 (Python Code)
Max Activating Token Index: 940
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2500. Min Act: -0.1700
Data Index: 1549213 (Python Code)
Max Activating Token Index: 465
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2969. Min Act: -0.1700
Data Index: 1395631 (Python Code)
Max Activating Token Index: 803
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2813. Min Act: -0.1700
Data Index: 305605 (C4 (Web Text))
Max Activating Token Index: 967
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2656. Min Act: -0.1700
Data Index: 1496733 (Python Code)
Max Activating Token Index: 145
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2500. Min Act: -0.1700
Data Index: 1647704 (Python Code)
Max Activating Token Index: 289
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2813. Min Act: -0.1700
Data Index: 1523396 (Python Code)
Max Activating Token Index: 784
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2656. Min Act: -0.1700
Data Index: 534422 (C4 (Web Text))
Max Activating Token Index: 628
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2500. Min Act: -0.1700
Data Index: 1576190 (Python Code)
Max Activating Token Index: 693
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2500. Min Act: -0.1700
Data Index: 1016165 (C4 (Web Text))
Max Activating Token Index: 679
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.2344. Min Act: -0.1700
Data Index: 722316 (C4 (Web Text))
Max Activating Token Index: 251
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.1875. Min Act: -0.1700
Data Index: 571872 (C4 (Web Text))
Max Activating Token Index: 281
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.1719. Min Act: -0.1700
Data Index: 1484338 (Python Code)
Max Activating Token Index: 431
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.1875. Min Act: -0.1700
Data Index: 933335 (C4 (Web Text))
Max Activating Token Index: 998
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.4844. Min Range: -4.4844
Max Act: 4.1719. Min Act: -0.1700
Data Index: 1365698 (Python Code)
Max Activating Token Index: 528
Click toggle to see full text
Truncated
Full Text #19