© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 7561
    Prev
    Next
    INDEX
    Explanations

    German word endings

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

    The neuron strongly activates on short character‐sequence fragments (like “eng,” “iso,” “dorf,” “satz,” “folge,” “end,” “dos,” “vo,” “lan,” “lg,” “ing,” “eld,” etc.) that occur as interior or terminal morphemes in compound names and technical or place‐name words.

    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
     of
    -2.36
     on
    -2.05
    🥂
    -1.89
    ый
    -1.84
     to
    -1.81
    🤯
    -1.73
     by
    -1.71
     are
    -1.70
    in
    -1.70
    💞
    -1.69
    POSITIVE LOGITS
    苼
    2.25
    眎
    2.14
     kaplama
    2.02
     eſſ
    1.94
    で使
    1.91
     galinha
    1.90
     pomys
    1.89
    Ꮙ
    1.88
     ſie
    1.86
    舠
    1.84
    Activations Density 0.012%

    No Known Activations