© 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. 22-GEMMASCOPE-RES-131K
    7. 65584
    Prev
    Next
    INDEX
    Explanations

    key markers and questions

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

    The neuron fires on standalone numeric tokens (e.g. the length or count markers like “332,” “92.5,” “350,” 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_22/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
     Português
    -0.96
    にほんブログ村
    -0.92
    েই
    -0.88
    Sinopse
    -0.85
    PrototypeOf
    -0.85
     cím
    -0.84
     Ensino
    -0.83
     сливо
    -0.82
    nění
    -0.82
     BitmapFactory
    -0.80
    POSITIVE LOGITS
    vided
    0.97
     yzf
    0.88
     why
    0.87
    したのが
    0.87
     hôtes
    0.87
     пригла
    0.84
    ;<
    0.84
     дополнитель
    0.83
     retras
    0.83
     dlatego
    0.82
    Activations Density 0.028%

    No Known Activations