© 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. 112879
    Prev
    Next
    INDEX
    Explanations

    what is being done

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

    The neuron activates on various inflected forms of the verb “to be” (e.g. am, be, was, were, are).

    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
    {
    -2.09
    觧
    -1.90
     only
    -1.82
    是吗
    -1.74
     тільки
    -1.66
    你这个
    -1.66
    哈哈哈哈
    -1.64
    诶
    -1.63
     Как
    -1.62
     pè
    -1.61
    POSITIVE LOGITS
     which
    1.98
     navegador
    1.55
    o
    1.52
     verschill
    1.45
     ?”
    1.45
    𖥧
    1.42
     تلك
    1.42
     whatnot
    1.41
     クラス
    1.38
    gestone
    1.36
    Activations Density 0.017%

    No Known Activations