© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 7536
    Prev
    Next
    INDEX
    Explanations

    self-deprecating or negative statements

    np_acts-logits-general · gemini-2.5-flash-lite

    simulating prejudiced viewpoints

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ；
    0.47
    Ⅰ
    0.46
         
    0.45
    ‥
    0.44
               
    0.44
              
    0.44
             
    0.43
    開始
    0.43
                                  
    0.41
    が見
    0.41
    POSITIVE LOGITS
     assh
    0.85
     misog
    0.85
     shitty
    0.82
     asshole
    0.82
     fucking
    0.79
     merda
    0.76
     idiots
    0.75
     fucked
    0.73
     shit
    0.70
     bigotry
    0.68
    Activations Density 0.122%

    No Known Activations