© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 71365
    Prev
    Next
    INDEX
    Explanations

    WordPress customizations and code comments

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

    The neuron fires strongly on standalone code identifiers—especially library, module, or API names (e.g. “Pattern Lab,” “Jest Snapshot,” “WordPress,” “withTheme”).

    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
     asegurarse
    -1.18
    bollah
    -1.12
     notwendig
    -1.12
    manız
    -1.11
     esquecer
    -1.10
     muligt
    -1.09
     endroit
    -1.07
     complicada
    -1.03
     OnInit
    -1.03
     It
    -1.03
    POSITIVE LOGITS
     the
    1.70
    ,),
    1.52
     mereka
    1.43
    `)
    1.37
     ellos
    1.34
    )`
    1.30
    elance
    1.29
    !」
    1.23
    _),
    1.23
    \}=
    1.20
    Activations Density 0.099%

    No Known Activations