© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 44618
    Prev
    Next
    INDEX
    Explanations

    code file names and definitions

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

    The neuron is effectively dead—it never activates on any token.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     allow
    -1.95
    }
    -1.90
     if
    -1.90
     as
    -1.88
    $
    -1.77
    ↵↵↵
    -1.75
     striving
    -1.69
     after
    -1.64
     strives
    -1.61
    You
    -1.59
    POSITIVE LOGITS
    sandalia
    1.85
    ZZLE
    1.78
    EMENTS
    1.76
    various
    1.74
     evidentemente
    1.68
     prácticamente
    1.64
    CurtirCurtir
    1.63
    
    1.63
     sorprender
    1.63
    訁
    1.61
    Activations Density 0.002%

    No Known Activations