Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1517 in Layer 2
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 4.3064. Min Range: -4.3064
Max Act: 4.3064. Min Act: -0.1700
Data Index: 423811 (C4 (Web Text))
Max Activating Token Index: 460
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.3064. Min Range: -4.3064
Max Act: 4.2127. Min Act: -0.1700
Data Index: 924684 (C4 (Web Text))
Max Activating Token Index: 64
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.3064. Min Range: -4.3064
Max Act: 4.0876. Min Act: -0.1700
Data Index: 123365 (C4 (Web Text))
Max Activating Token Index: 339
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.9938. Min Act: -0.1700
Data Index: 441689 (C4 (Web Text))
Max Activating Token Index: 559
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.8687. Min Act: -0.1700
Data Index: 494774 (C4 (Web Text))
Max Activating Token Index: 818
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.8687. Min Act: -0.1700
Data Index: 982792 (C4 (Web Text))
Max Activating Token Index: 776
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.8687. Min Act: -0.1700
Data Index: 155959 (C4 (Web Text))
Max Activating Token Index: 637
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.8062. Min Act: -0.1700
Data Index: 756497 (C4 (Web Text))
Max Activating Token Index: 402
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.8062. Min Act: -0.1700
Data Index: 774393 (C4 (Web Text))
Max Activating Token Index: 313
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.8062. Min Act: -0.1700
Data Index: 1111731 (C4 (Web Text))
Max Activating Token Index: 865
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.7749. Min Act: -0.1700
Data Index: 507285 (C4 (Web Text))
Max Activating Token Index: 586
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.8062. Min Act: -0.1700
Data Index: 771780 (C4 (Web Text))
Max Activating Token Index: 323
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.7123. Min Act: -0.1700
Data Index: 679476 (C4 (Web Text))
Max Activating Token Index: 795
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.7436. Min Act: -0.1700
Data Index: 100855 (C4 (Web Text))
Max Activating Token Index: 104
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.6810. Min Act: -0.1700
Data Index: 694484 (C4 (Web Text))
Max Activating Token Index: 279
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.6810. Min Act: -0.1700
Data Index: 1006005 (C4 (Web Text))
Max Activating Token Index: 329
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.7123. Min Act: -0.1700
Data Index: 87422 (C4 (Web Text))
Max Activating Token Index: 685
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.6967. Min Act: -0.1700
Data Index: 875904 (C4 (Web Text))
Max Activating Token Index: 101
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.6810. Min Act: -0.1700
Data Index: 315840 (C4 (Web Text))
Max Activating Token Index: 28
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.3064. Min Range: -4.3064
Max Act: 3.6028. Min Act: -0.1700
Data Index: 476937 (C4 (Web Text))
Max Activating Token Index: 830
Click toggle to see full text
Truncated
Full Text #19