© 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. 10-GEMMASCOPE-RES-131K
    7. 60088
    Prev
    Next
    INDEX
    Explanations

    number of times

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

    The neuron specializes in spotting phrases that state or ask about numerical limits or capacities—e.g. “the maximum number of X that can be Y.”

    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
    }
    -1.91
     jaket
    -1.54
     also
    -1.48
     quelque
    -1.48
     другое
    -1.47
    ”?
    -1.46
    ";
    -1.45
    皆さま
    -1.45
     cuanto
    -1.44
     peluche
    -1.43
    POSITIVE LOGITS
     from
    1.78
     for
    1.63
     the
    1.63
     on
    1.58
     each
    1.55
     to
    1.55
    ditionally
    1.51
     this
    1.44
     during
    1.38
    laundry
    1.38
    Activations Density 0.080%

    No Known Activations