Model: GELU Model: 3 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 678 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')
Text #0
Max Range: 4.1305. Min Range: -4.1305
Max Act: 4.1305. Min Act: -0.1700
Data Index: 1316871 (C4 (Web Text))
Max Activating Token Index: 59
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.1305. Min Range: -4.1305
Max Act: 4.0993. Min Act: -0.1700
Data Index: 1532376 (Python Code)
Max Activating Token Index: 256
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.1305. Min Range: -4.1305
Max Act: 4.0680. Min Act: -0.1700
Data Index: 679281 (C4 (Web Text))
Max Activating Token Index: 910
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.1305. Min Range: -4.1305
Max Act: 4.0680. Min Act: -0.1700
Data Index: 1450070 (Python Code)
Max Activating Token Index: 230
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.1305. Min Range: -4.1305
Max Act: 4.0680. Min Act: -0.1700
Data Index: 1099114 (C4 (Web Text))
Max Activating Token Index: 386
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.1305. Min Range: -4.1305
Max Act: 4.0680. Min Act: -0.1700
Data Index: 1497546 (Python Code)
Max Activating Token Index: 741
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.1305. Min Range: -4.1305
Max Act: 4.0055. Min Act: -0.1700
Data Index: 1548919 (Python Code)
Max Activating Token Index: 919
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.1305. Min Range: -4.1305
Max Act: 4.0055. Min Act: -0.1700
Data Index: 1006687 (C4 (Web Text))
Max Activating Token Index: 717
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9429. Min Act: -0.1700
Data Index: 1481839 (Python Code)
Max Activating Token Index: 897
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9742. Min Act: -0.1700
Data Index: 1623660 (Python Code)
Max Activating Token Index: 82
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9742. Min Act: -0.1700
Data Index: 1471507 (Python Code)
Max Activating Token Index: 449
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9742. Min Act: -0.1700
Data Index: 1485211 (Python Code)
Max Activating Token Index: 365
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9429. Min Act: -0.1700
Data Index: 865724 (C4 (Web Text))
Max Activating Token Index: 718
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9742. Min Act: -0.1700
Data Index: 1307663 (C4 (Web Text))
Max Activating Token Index: 844
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9429. Min Act: -0.1700
Data Index: 1158119 (C4 (Web Text))
Max Activating Token Index: 690
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9742. Min Act: -0.1700
Data Index: 599244 (C4 (Web Text))
Max Activating Token Index: 729
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9429. Min Act: -0.1700
Data Index: 1393019 (Python Code)
Max Activating Token Index: 232
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9429. Min Act: -0.1700
Data Index: 1629976 (Python Code)
Max Activating Token Index: 963
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9117. Min Act: -0.1700
Data Index: 1449072 (Python Code)
Max Activating Token Index: 537
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.1305. Min Range: -4.1305
Max Act: 3.9117. Min Act: -0.1700
Data Index: 1470295 (Python Code)
Max Activating Token Index: 115
Click toggle to see full text
Truncated
Full Text #19