Model: GPT-2 Xl: 48 Layers, 6400 Neurons per Layer
Dataset: Open Web Text
Neuron 1356 in Layer 1
Transformer Lens Loading: HookedTransformer.from_pretrained('gpt2-xl')
Text #0
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.6489. Min Act: -0.1697
Data Index: 7930010 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.6140. Min Act: -0.1700
Data Index: 8823111 (Open Web Text)
Max Activating Token Index: 3
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5965. Min Act: -0.1700
Data Index: 3219030 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5791. Min Act: -0.1700
Data Index: 6538449 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5616. Min Act: -0.1693
Data Index: 5449464 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5265. Min Act: -0.1700
Data Index: 2217088 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5265. Min Act: -0.1700
Data Index: 2574199 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5441. Min Act: -0.1700
Data Index: 4149515 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5265. Min Act: -0.1700
Data Index: 7667890 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5265. Min Act: -0.1678
Data Index: 3979561 (Open Web Text)
Max Activating Token Index: 3
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.4914. Min Act: -0.1700
Data Index: 6113700 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5090. Min Act: -0.1700
Data Index: 2434781 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5090. Min Act: -0.1700
Data Index: 5794015 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5090. Min Act: -0.1700
Data Index: 4354982 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5090. Min Act: -0.1700
Data Index: 8332658 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.5090. Min Act: -0.1700
Data Index: 4507653 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.4914. Min Act: -0.1700
Data Index: 7645717 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.4914. Min Act: -0.1700
Data Index: 3905119 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.4739. Min Act: -0.1700
Data Index: 4105934 (Open Web Text)
Max Activating Token Index: 2
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 1.6489. Min Range: -1.6489
Max Act: 1.4914. Min Act: -0.1700
Data Index: 4970246 (Open Web Text)
Max Activating Token Index: 4
Click toggle to see full text
Truncated
Full Text #19