Model: GELU Model: 4 Layers, 2048 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 2027 in Layer 0
Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')
Text #0
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.7787. Min Act: -0.1700
Data Index: 40564 (C4 (Web Text))
Max Activating Token Index: 948
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.7308. Min Act: -0.1700
Data Index: 1379194 (Python Code)
Max Activating Token Index: 122
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6989. Min Act: -0.1700
Data Index: 529656 (C4 (Web Text))
Max Activating Token Index: 247
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.7149. Min Act: -0.1700
Data Index: 1396006 (Python Code)
Max Activating Token Index: 23
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6989. Min Act: -0.1700
Data Index: 197834 (C4 (Web Text))
Max Activating Token Index: 991
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6828. Min Act: -0.1700
Data Index: 1539645 (Python Code)
Max Activating Token Index: 764
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6828. Min Act: -0.1700
Data Index: 717430 (C4 (Web Text))
Max Activating Token Index: 165
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6989. Min Act: -0.1700
Data Index: 986515 (C4 (Web Text))
Max Activating Token Index: 814
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6828. Min Act: -0.1700
Data Index: 507956 (C4 (Web Text))
Max Activating Token Index: 325
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6668. Min Act: -0.1700
Data Index: 1436896 (Python Code)
Max Activating Token Index: 49
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6828. Min Act: -0.1700
Data Index: 1364254 (Python Code)
Max Activating Token Index: 218
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6668. Min Act: -0.1700
Data Index: 1272680 (C4 (Web Text))
Max Activating Token Index: 762
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6828. Min Act: -0.1700
Data Index: 323014 (C4 (Web Text))
Max Activating Token Index: 122
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6668. Min Act: -0.1700
Data Index: 807498 (C4 (Web Text))
Max Activating Token Index: 943
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6508. Min Act: -0.1700
Data Index: 828096 (C4 (Web Text))
Max Activating Token Index: 316
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6668. Min Act: -0.1700
Data Index: 62738 (C4 (Web Text))
Max Activating Token Index: 847
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6668. Min Act: -0.1700
Data Index: 167669 (C4 (Web Text))
Max Activating Token Index: 726
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6347. Min Act: -0.1700
Data Index: 73215 (C4 (Web Text))
Max Activating Token Index: 257
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6508. Min Act: -0.1700
Data Index: 1047971 (C4 (Web Text))
Max Activating Token Index: 756
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 2.7787. Min Range: -2.7787
Max Act: 2.6668. Min Act: -0.1700
Data Index: 100330 (C4 (Web Text))
Max Activating Token Index: 888
Click toggle to see full text
Truncated
Full Text #19