© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 116217
    Prev
    Next
    INDEX
    Explanations

    property tax parcels

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

    This neuron detects the key tokens in the question header—specifically the words naming the object of inquiry (e.g. “a,” “land,” “survey,” “for,” “my,” “house”) in a “how to locate”-style property question.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     cộng
    -0.76
    epsfig
    -0.71
    mies
    -0.69
     ungrateful
    -0.68
     inaugurate
    -0.66
     inaugurated
    -0.66
    っちり
    -0.65
    agence
    -0.64
    ewear
    -0.64
     supplier
    -0.64
    POSITIVE LOGITS
     parcel
    2.55
    Parcel
    2.23
     parcels
    2.16
    parcel
    2.13
     Parcel
    2.06
     tax
    2.00
     Tax
    1.89
    Tax
    1.89
    tax
    1.64
     Parc
    1.56
    Activations Density 0.026%

    No Known Activations