Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    tables summarizing differences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
             
    1.40
              
    1.35
               
    1.34
                 
    1.30
                
    1.24
                  
    1.18
    1.15
                   
    1.12
                     
    1.10
            
    1.07
    POSITIVE LOGITS
     assignment
    0.77
    --");
    0.76
    --',
    0.73
     playthrough
    0.73
    )}$,
    0.72
    )..
    0.71
     assignments
    0.71
    ..)
    0.71
     streamline
    0.69
    }){
    0.67
    Act Density 0.017%

    No Known Activations