© 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. 2201
    Prev
    Next
    INDEX
    Explanations

    fraction expression

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

    The neuron primarily detects numeric tokens and symbols used in mathematical or scientific expressions (e.g. numbers, decimals, multipliers like “×,” and variance notation).

    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
     There
    0.63
     This
    0.61
    ⅔
    0.59
    ¾
    0.59
     ½
    0.58
     ¾
    0.57
     Therefore
    0.56
     thys
    0.55
     ด้วย
    0.54
     These
    0.53
    POSITIVE LOGITS
    }{\
    0.61
    gher
    0.54
    상
    0.54
    了一下
    0.53
    INDIA
    0.52
    俄
    0.52
    msqrt
    0.52
    )}(\
    0.52
    いろ
    0.51
    会
    0.51
    Activations Density 0.003%

    No Known Activations