Model: SoLU Model: 8 Layers, 4096 Neurons per Layer
Dataset: 80% C4 (Web Text) and 20% Python Code
Neuron 1989 in Layer 4
Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')
Text #0
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.8488. Min Act: -0.0001
Data Index: 960096 (C4 (Web Text))
Max Activating Token Index: 944
Click toggle to see full text
Truncated
Full Text #0
Text #1
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.6966. Min Act: -0.0001
Data Index: 117226 (C4 (Web Text))
Max Activating Token Index: 338
Click toggle to see full text
Truncated
Full Text #1
Text #2
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.7596. Min Act: -0.0001
Data Index: 806688 (C4 (Web Text))
Max Activating Token Index: 364
Click toggle to see full text
Truncated
Full Text #2
Text #3
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.5748. Min Act: -0.0001
Data Index: 680157 (C4 (Web Text))
Max Activating Token Index: 755
Click toggle to see full text
Truncated
Full Text #3
Text #4
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.5358. Min Act: -0.0001
Data Index: 195932 (C4 (Web Text))
Max Activating Token Index: 637
Click toggle to see full text
Truncated
Full Text #4
Text #5
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.4464. Min Act: -0.0001
Data Index: 1118611 (C4 (Web Text))
Max Activating Token Index: 465
Click toggle to see full text
Truncated
Full Text #5
Text #6
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.5010. Min Act: -0.0001
Data Index: 43669 (C4 (Web Text))
Max Activating Token Index: 203
Click toggle to see full text
Truncated
Full Text #6
Text #7
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.4465. Min Act: -0.0001
Data Index: 1212851 (C4 (Web Text))
Max Activating Token Index: 243
Click toggle to see full text
Truncated
Full Text #7
Text #8
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.3868. Min Act: -0.0001
Data Index: 395577 (C4 (Web Text))
Max Activating Token Index: 1013
Click toggle to see full text
Truncated
Full Text #8
Text #9
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.4081. Min Act: -0.0001
Data Index: 115219 (C4 (Web Text))
Max Activating Token Index: 219
Click toggle to see full text
Truncated
Full Text #9
Text #10
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.3724. Min Act: -0.0001
Data Index: 706205 (C4 (Web Text))
Max Activating Token Index: 836
Click toggle to see full text
Truncated
Full Text #10
Text #11
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.3907. Min Act: -0.0001
Data Index: 711328 (C4 (Web Text))
Max Activating Token Index: 441
Click toggle to see full text
Truncated
Full Text #11
Text #12
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.3911. Min Act: -0.0001
Data Index: 607832 (C4 (Web Text))
Max Activating Token Index: 344
Click toggle to see full text
Truncated
Full Text #12
Text #13
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.3990. Min Act: -0.0001
Data Index: 470539 (C4 (Web Text))
Max Activating Token Index: 677
Click toggle to see full text
Truncated
Full Text #13
Text #14
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.2827. Min Act: -0.0001
Data Index: 293147 (C4 (Web Text))
Max Activating Token Index: 556
Click toggle to see full text
Truncated
Full Text #14
Text #15
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.4075. Min Act: -0.0001
Data Index: 1244689 (C4 (Web Text))
Max Activating Token Index: 26
Click toggle to see full text
Truncated
Full Text #15
Text #16
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.3700. Min Act: -0.0001
Data Index: 668868 (C4 (Web Text))
Max Activating Token Index: 388
Click toggle to see full text
Truncated
Full Text #16
Text #17
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.4094. Min Act: -0.0001
Data Index: 804588 (C4 (Web Text))
Max Activating Token Index: 255
Click toggle to see full text
Truncated
Full Text #17
Text #18
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.3459. Min Act: -0.0001
Data Index: 808714 (C4 (Web Text))
Max Activating Token Index: 341
Click toggle to see full text
Truncated
Full Text #18
Text #19
Max Range: 1.8488. Min Range: -1.8488
Max Act: 1.2616. Min Act: -0.0001
Data Index: 582847 (C4 (Web Text))
Max Activating Token Index: 209
Click toggle to see full text
Truncated
Full Text #19