© 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. 22-GEMMASCOPE-RES-131K
    7. 25839
    Prev
    Next
    INDEX
    Explanations

    without discrimination

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

    This neuron lights up on relatively rare or specialized tokens, i.e. low‐frequency words (and punctuation) that tend to occur in domain‐specific or technical contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     Gehalt
    -0.96
    度の
    -0.91
     Descripción
    -0.90
    ﾍ
    -0.90
     Tempat
    -0.89
    ֢
    -0.87
    bảng
    -0.87
    煸
    -0.84
    =[];
    -0.83
     vụ
    -0.82
    POSITIVE LOGITS
    ͌
    1.05
     the
    0.98
    žená
    0.96
     they
    0.92
     firebaseConfig
    0.88
    IAC
    0.86
     coûts
    0.85
    
    0.85
    indows
    0.85
     scherm
    0.85
    Activations Density 0.062%

    No Known Activations