© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 118843
    Prev
    Next
    INDEX
    Explanations

    nearest integer

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

    The neuron activates on superlative and comparative proximity terms (e.g. “nearest,” “closest,” “best”) indicating an extremal/location comparison.

    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
    𐍆
    -1.71
    mmmm
    -1.67
    땃
    -1.65
    菝
    -1.64
     ❓
    -1.62
    ኹ
    -1.61
     bof
    -1.60
     Genel
    -1.58
     内衣
    -1.58
     ork
    -1.55
    POSITIVE LOGITS
     that
    2.13
     was
    1.98
     either
    1.93
     I
    1.79
     or
    1.77
     wasn
    1.77
    3
    1.74
     mention
    1.73
    2
    1.68
    —
    1.63
    Activations Density 0.021%

    No Known Activations