© 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. 128795
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on fragments of technical biochemical or pharmaceutical modifiers (e.g. parts of “reversibly activatable,” “cell-penetrating,” or “pH-sensitive” terms).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    Hotel ID

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     prí
    -0.85
    ɐ
    -0.84
    ولی
    -0.83
    ittä
    -0.82
    šie
    -0.82
    疙
    -0.81
    ⼈
    -0.81
     arrep
    -0.80
     incluso
    -0.79
     médica
    -0.77
    POSITIVE LOGITS
     fenomeno
    0.93
     powerhouse
    0.88
    atissimo
    0.87
     realising
    0.82
     определить
    0.82
    商品
    0.79
    JA
    0.78
    当
    0.75
     recognising
    0.75
    ثر
    0.75
    Activations Density 0.106%

    No Known Activations