© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 85747
    Prev
    Next
    INDEX
    Explanations

    dependence coefficient

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

    The neuron fires on uppercase technical acronyms and protocol names (e.g. ODBC, OBD, PBKDF2, SOCKS).

    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
    er
    -2.16
    </em>
    -1.78
    a
    -1.72
    ."
    -1.71
     وت
    -1.61
     do
    -1.59
     del
    -1.57
    ings
    -1.56
    把它
    -1.56
    </strong>
    -1.55
    POSITIVE LOGITS
    鮋
    1.80
     sayap
    1.73
     menyen
    1.66
     terbang
    1.61
    鬭
    1.59
     menyerang
    1.59
    鏵
    1.57
     ledig
    1.55
     weirdly
    1.55
     Eropa
    1.54
    Activations Density 0.016%

    No Known Activations