© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2135
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions

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

    The neuron fires on math-instruction and algebraic terminology (words associated with simplifying expressions, radical operations, equations, and other math problem-solving language).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     létre
    0.57
     curva
    0.57
     trama
    0.54
     admis
    0.54
     incr
    0.53
    懺
    0.53
     besch
    0.51
     informacija
    0.50
    감독
    0.50
    获得了
    0.49
    POSITIVE LOGITS
     Simplify
    0.94
     monomials
    0.92
     Simplifying
    0.91
     выражение
    0.89
    Simplify
    0.89
    simplify
    0.88
     Expressions
    0.86
     simplifying
    0.86
    -(-
    0.84
     Multiplication
    0.82
    Activations Density 0.014%

    No Known Activations