© 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. 66915
    Prev
    Next
    INDEX
    Explanations

    evaluating a status

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

    The neuron activates on mentions of various “status” assessments (e.g., methylation status, pollution status, oxidative status, nutritional status, susceptibility status, etc.).

    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
     to
    -2.81
    unno
    -1.68
     verschillende
    -1.67
    actéristi
    -1.65
     the
    -1.62
    nade
    -1.60
    あぁ
    -1.56
    ふと
    -1.55
    えっ
    -1.55
    xvii
    -1.53
    POSITIVE LOGITS
     if
    2.06
    6
    1.98
    3
    1.77
     or
    1.64
    4
    1.52
     at
    1.50
     even
    1.49
     suya
    1.47
     sleek
    1.45
     crazy
    1.45
    Activations Density 0.098%

    No Known Activations