© 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. 67240
    Prev
    Next
    INDEX
    Explanations

    loved one or ones

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

    The neuron fires strongly on the phrase “loved one” (and its plural), i.e. references to someone’s “loved one(s).”

    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
    謖
    -2.23
    萇
    -1.95
    髡
    -1.94
    ."
    -1.88
    </strong>
    -1.86
    ";//
    -1.84
    錸
    -1.80
    '));
    -1.80
     kjer
    -1.80
    ньому
    -1.79
    POSITIVE LOGITS
    d
    2.27
    ’
    2.08
     They
    2.08
    在他
    1.95
    在她
    1.89
    "
    1.88
     consequent
    1.88
    -
    1.84
    di
    1.83
    始めて
    1.77
    Activations Density 0.002%

    No Known Activations