© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 61142
    Prev
    Next
    INDEX
    Explanations

    $x^2$ and $y^2$ expressions

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

    This neuron activates on numeric tokens (e.g. integers or decimals) in mathematical expressions—that is, it detects literal numbers.

    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
    or
    -1.72
    ramientas
    -1.66
    紧急
    -1.58
    aterin
    -1.56
    阴影
    -1.48
    s
    -1.46
     региона
    -1.45
     Ortschaft
    -1.41
    ників
    -1.41
    phée
    -1.40
    POSITIVE LOGITS
    -------
    
    2.02
     conquistar
    1.77
    --
    
    1.75
    THEY
    1.65
     немає
    1.62
    ----------
    
    1.61
    There
    1.60
    If
    1.59
    جوا
    1.59
     betreft
    1.59
    Activations Density 0.033%

    No Known Activations