Model: GELU Model: 2 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 2014 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')
Text #0
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.9030. Min Act: -0.1700
Data Index: 448837 (C4 (Web Text))
Max Activating Token Index: 47
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.8092. Min Act: -0.1700
Data Index: 613058 (C4 (Web Text))
Max Activating Token Index: 51
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.7780. Min Act: -0.1700
Data Index: 1093678 (C4 (Web Text))
Max Activating Token Index: 43
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.7780. Min Act: -0.1700
Data Index: 858430 (C4 (Web Text))
Max Activating Token Index: 53
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.7155. Min Act: -0.1700
Data Index: 644604 (C4 (Web Text))
Max Activating Token Index: 34
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.7155. Min Act: -0.1700
Data Index: 1188761 (C4 (Web Text))
Max Activating Token Index: 53
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.6842. Min Act: -0.1700
Data Index: 1020016 (C4 (Web Text))
Max Activating Token Index: 43
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.6530. Min Act: -0.1700
Data Index: 224426 (C4 (Web Text))
Max Activating Token Index: 53
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.6842. Min Act: -0.1700
Data Index: 269125 (C4 (Web Text))
Max Activating Token Index: 38
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.6217. Min Act: -0.1700
Data Index: 283470 (C4 (Web Text))
Max Activating Token Index: 49
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.5905. Min Act: -0.1700
Data Index: 917930 (C4 (Web Text))
Max Activating Token Index: 41
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.5905. Min Act: -0.1700
Data Index: 999378 (C4 (Web Text))
Max Activating Token Index: 399
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.5905. Min Act: -0.1700
Data Index: 612294 (C4 (Web Text))
Max Activating Token Index: 34
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.6217. Min Act: -0.1700
Data Index: 905219 (C4 (Web Text))
Max Activating Token Index: 41
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.6217. Min Act: -0.1700
Data Index: 1153576 (C4 (Web Text))
Max Activating Token Index: 31
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.5905. Min Act: -0.1700
Data Index: 1351645 (C4 (Web Text))
Max Activating Token Index: 41
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.5592. Min Act: -0.1700
Data Index: 682368 (C4 (Web Text))
Max Activating Token Index: 38
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.5592. Min Act: -0.1700
Data Index: 644803 (C4 (Web Text))
Max Activating Token Index: 31
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.5905. Min Act: -0.1700
Data Index: 1259090 (C4 (Web Text))
Max Activating Token Index: 155
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 4.9030. Min Range: -4.9030
Max Act: 4.5592. Min Act: -0.1700
Data Index: 352351 (C4 (Web Text))
Max Activating Token Index: 45
Click toggle to see full text
Truncated
Full Text #19