© 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. 15425
    Prev
    Next
    INDEX
    Explanations

    formula involving fractions, sums, or integrals

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

    This neuron activates on TeX/LaTeX math commands (backslash-initiated tokens) 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_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
    }$
    -1.58
    laine
    -1.56
     Desember
    -1.53
    thday
    -1.52
    Hrs
    -1.50
    ных
    -1.49
    razine
    -1.48
     Februari
    -1.46
    などの
    -1.45
    🌲
    -1.45
    POSITIVE LOGITS
    えっ
    1.65
    様な
    1.55
    にします
    1.54
    を行います
    1.52
    に入れる
    1.47
     необхідно
    1.45
    うれしい
    1.44
    総評
    1.44
    出于
    1.44
     günstiger
    1.41
    Activations Density 0.007%

    No Known Activations