© 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. 69978
    Prev
    Next
    INDEX
    Explanations

    submitting proposals

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

    This neuron is triggered by the initial “The” (and other sentence‐opening words) that start new sentences or paragraphs.

    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
     then
    -2.09
     різних
    -1.88
     Años
    -1.80
    瘣
    -1.79
     Ajoutez
    -1.77
     Informações
    -1.76
     ""],
    -1.73
     Besoin
    -1.65
    getSystemService
    -1.63
     Entonces
    -1.62
    POSITIVE LOGITS
     is
    2.59
     will
    1.91
     in
    1.77
     from
    1.73
    !
    1.68
     its
    1.60
     даже
    1.59
     "
    1.57
     у
    1.57
     According
    1.52
    Activations Density 0.052%

    No Known Activations