© 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. 40054
    Prev
    Next
    INDEX
    Explanations

    case, 9, "A", pseud, strictly

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

    The neuron fires on high‐level document headings or metadata markers (e.g. titles like “How to…,” section labels like “ABSTRACT,” salutations like “Dear Editor,” or docstring delimiters).

    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
     impecable
    -1.94
    儈
    -1.76
     taba
    -1.76
     parfaitement
    -1.70
    一口气
    -1.69
     noblesse
    -1.67
    蚡
    -1.67
    嘁
    -1.66
     maso
    -1.66
    犛
    -1.66
    POSITIVE LOGITS
     in
    2.75
    1
    2.63
     that
    2.20
    2
    2.20
     at
    2.14
    3
    2.08
    4
    2.03
    8
    2.03
    }
    1.98
     the
    1.91
    Activations Density 0.000%

    No Known Activations