© 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. 67548
    Prev
    Next
    INDEX
    Explanations

    Chabad, L’CHAIM, January, Education

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

    The neuron strongly lights up on capitalized subword pieces—especially the prefixes of proper names, acronyms, and other title‐case tokens.

    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
     a
    -1.63
    </h3>
    -1.62
     need
    -1.61
    </h1>
    -1.60
    </u>
    -1.55
     are
    -1.55
    一个
    -1.52
     vorhand
    -1.39
    }
    -1.38
     就
    -1.37
    POSITIVE LOGITS
    涐
    1.82
     parachoque
    1.67
    sandália
    1.61
    اشتم
    1.58
     maquina
    1.55
     充電
    1.55
     playera
    1.52
     夜景
    1.52
    ㍑
    1.52
     coreana
    1.52
    Activations Density 0.007%

    No Known Activations