© 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. 65062
    Prev
    Next
    INDEX
    Explanations

    king, queen, ruler

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

    The neuron fires strongly on proper nouns and named entities (people, mythological figures, animals, etc.).

    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
    ּוֹ
    -2.70
    ִּ
    -2.09
    וּ
    -2.00
     πρ
    -1.95
    ֶּ
    -1.95
     бампер
    -1.81
    ַּ
    -1.81
     брюки
    -1.49
    ֵּ
    -1.48
    Моя
    -1.48
    POSITIVE LOGITS
    ׇ
    2.00
     گیاه
    1.80
     to
    1.76
     after
    1.61
     by
    1.59
     from
    1.55
     his
    1.51
     he
    1.45
    ὴν
    1.45
     him
    1.42
    Activations Density 0.048%

    No Known Activations