© 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. 48292
    Prev
    Next
    INDEX
    Explanations

    mathematical variable m

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

    The neuron fires on the letter “m” when it appears as a mathematical variable or parameter in formulas (especially in LaTeX‐style math).

    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
    cadre
    -1.98
    ldk
    -1.95
    theres
    -1.77
    schermata
    -1.77
    ktm
    -1.75
    rmk
    -1.73
     innovations
    -1.72
     oba
    -1.71
     ();
    -1.66
     respectivos
    -1.66
    POSITIVE LOGITS
    1
    2.55
    ),
    2.36
    5
    2.33
    7
    2.22
    I
    2.22
    4
    2.19
    6
    2.08
    3
    2.03
    на
    1.99
    8
    1.94
    Activations Density 0.014%

    No Known Activations