© 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. 73684
    Prev
    Next
    INDEX
    Explanations

    JSON keys and properties

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

    This neuron identifies object‐literal property names (identifiers immediately followed by a colon) in code.

    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
     of
    -3.09
     who
    -2.08
    趿
    -1.85
     also
    -1.60
    </em>
    -1.56
     located
    -1.55
     involved
    -1.55
     announced
    -1.55
     whose
    -1.54
     offering
    -1.54
    POSITIVE LOGITS
     herido
    1.69
     revelar
    1.55
    Perhaps
    1.51
    Although
    1.50
     sırasında
    1.49
     diarrhea
    1.44
    although
    1.44
     }))
    1.42
    XYGEN
    1.41
     sentado
    1.41
    Activations Density 0.057%

    No Known Activations