© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 11337
    Prev
    Next
    INDEX
    Explanations

    weight and kernel parameters

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

    The neuron responds to technical machine‐learning parameter and hyperparameter terms (like “weight,” “gradient,” “convolution,” etc.) often paired with numeric values.

    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
    if
    -2.03
    ate
    -1.97
    ly
    -1.77
    ."
    -1.77
    ',
    -1.66
    3
    -1.64
     if
    -1.64
     when
    -1.61
     can
    -1.60
    or
    -1.55
    POSITIVE LOGITS
    kệ
    1.82
    genodigd
    1.47
    ᯾
    1.46
    Ꮉ
    1.43
    
    1.41
     downloaden
    1.38
    szt
    1.37
    𖧵
    1.37
    bảng
    1.35
     هناك
    1.34
    Activations Density 0.005%

    No Known Activations