© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14803
    Prev
    Next
    INDEX
    Explanations

    technical keywords and identifiers

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

    The neuron fires on programming‐style identifiers and API/library names (e.g. CamelCase class or constant tokens and all‐caps acronyms).

    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
     and
    -3.47
     "
    -1.74
     The
    -1.50
     my
    -1.48
     &
    -1.40
    <strong>
    -1.36
     their
    -1.36
    横に
    -1.34
     What
    -1.34
     I
    -1.27
    POSITIVE LOGITS
     褲
    1.57
    鏨
    1.56
     héro
    1.56
    hojas
    1.55
    θηκε
    1.54
    Penulis
    1.52
     sostu
    1.51
    pagina
    1.50
    tiefel
    1.48
    𝓜
    1.48
    Activations Density 0.092%

    No Known Activations