© 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. 22-GEMMASCOPE-RES-131K
    4. 70666
    Prev
    Next
    INDEX
    Explanations

    assigning values to attributes

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

    The neuron fires on parameter or attribute names (like “color,” “link,” “scale,” “image,” “Tag,” etc.) used in key–value settings across code and markup.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     will
    -1.46
     with
    -1.45
    無し
    -1.41
    為に
    -1.41
     this
    -1.38
     enough
    -1.34
     that
    -1.31
    ({
    
    -1.30
     which
    -1.27
    に入れ
    -1.27
    POSITIVE LOGITS
    や
    1.63
    などを
    1.53
    がなく
    1.50
    を
    1.46
    ftet
    1.45
    は
    1.44
    <bos>
    1.39
    Biblia
    1.38
    Manfaat
    1.37
    Ofic
    1.35
    Activations Density 0.001%

    No Known Activations