© 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. 34-GEMMASCOPE-RES-131K
    7. 3046
    Prev
    Next
    INDEX
    Explanations

    manifold M

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

    The neuron is triggered by LaTeX‐style mathematical notation, in particular variable names (like single uppercase letters), math operators, and structural tokens (e.g. “bullet”) used in formulas.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     rapidly
    -0.83
     Tuesday
    -0.79
     سریع
    -0.79
     Wednesday
    -0.77
     European
    -0.77
     changed
    -0.77
    سب
    -0.75
     and
    -0.75
     increased
    -0.75
     strongly
    -0.74
    POSITIVE LOGITS
    cong
    0.86
    ッグ
    0.83
    anything
    0.82
    øg
    0.82
    各个
    0.81
     atención
    0.81
    见的
    0.81
    它们的
    0.79
    iliger
    0.79
    presented
    0.79
    Activations Density 0.034%

    No Known Activations