© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 127996
    Prev
    Next
    INDEX
    Explanations

    identifying numbered editions

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

    The neuron detects references to numbered book or journal editions (e.g. ordinal markers like “2nd,” “Third,” “Fourth,” and words like “Edition” or “ed”).

    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
    鼉
    -1.92
    
    -1.80
     Improves
    -1.78
     Sugges
    -1.77
    鷽
    -1.72
     Integrations
    -1.71
    这一幕
    -1.69
     considerados
    -1.63
     fundament
    -1.62
     Prod
    -1.62
    POSITIVE LOGITS
    ↵
    2.16
    The
    1.93
     его
    1.83
    In
    1.74
    From
    1.69
     to
    1.60
     он
    1.59
    l
    1.56
    .”
    1.56
     –
    1.56
    Activations Density 0.006%

    No Known Activations