Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1120 in Layer 2
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 4.1181. Min Range: -4.1181
Max Act: 4.1181. Min Act: -0.1700
Data Index: 1271303 (C4 (Web Text))
Max Activating Token Index: 634
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.1181. Min Range: -4.1181
Max Act: 4.0868. Min Act: -0.1700
Data Index: 265916 (C4 (Web Text))
Max Activating Token Index: 117
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.1181. Min Range: -4.1181
Max Act: 4.0556. Min Act: -0.1700
Data Index: 886529 (C4 (Web Text))
Max Activating Token Index: 777
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.1181. Min Range: -4.1181
Max Act: 4.0556. Min Act: -0.1700
Data Index: 1547791 (Python Code)
Max Activating Token Index: 890
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9618. Min Act: -0.1700
Data Index: 103742 (C4 (Web Text))
Max Activating Token Index: 337
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9930. Min Act: -0.1700
Data Index: 208859 (C4 (Web Text))
Max Activating Token Index: 204
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9618. Min Act: -0.1700
Data Index: 345176 (C4 (Web Text))
Max Activating Token Index: 861
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9618. Min Act: -0.1700
Data Index: 644967 (C4 (Web Text))
Max Activating Token Index: 949
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9930. Min Act: -0.1700
Data Index: 1552551 (Python Code)
Max Activating Token Index: 650
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9618. Min Act: -0.1700
Data Index: 1458738 (Python Code)
Max Activating Token Index: 634
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9149. Min Act: -0.1700
Data Index: 212725 (C4 (Web Text))
Max Activating Token Index: 287
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9305. Min Act: -0.1700
Data Index: 964222 (C4 (Web Text))
Max Activating Token Index: 364
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9618. Min Act: -0.1700
Data Index: 438173 (C4 (Web Text))
Max Activating Token Index: 676
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9305. Min Act: -0.1700
Data Index: 765072 (C4 (Web Text))
Max Activating Token Index: 381
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9305. Min Act: -0.1700
Data Index: 218629 (C4 (Web Text))
Max Activating Token Index: 791
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9305. Min Act: -0.1700
Data Index: 1049428 (C4 (Web Text))
Max Activating Token Index: 663
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9149. Min Act: -0.1700
Data Index: 1290639 (C4 (Web Text))
Max Activating Token Index: 657
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.9149. Min Act: -0.1700
Data Index: 1470574 (Python Code)
Max Activating Token Index: 1019
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.8992. Min Act: -0.1700
Data Index: 1623660 (Python Code)
Max Activating Token Index: 393
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.1181. Min Range: -4.1181
Max Act: 3.8836. Min Act: -0.1700
Data Index: 1640798 (Python Code)
Max Activating Token Index: 690
Click toggle to see full text
Truncated
Full Text #19