© 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. 2324
    Prev
    Next
    INDEX
    Explanations

    This neuron flags numeric magnitude expressions, especially “up to”‐style limits or specifications (e.g. resolution, distance, maximum values).

    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
     Три
    0.60
    ْب
    0.60
    aziland
    0.58
    ẖ
    0.58
    𝑛
    0.57
    ｌ
    0.56
     過
    0.55
     Incorporation
    0.55
     再
    0.54
     كيل
    0.54
    POSITIVE LOGITS
     wannan
    0.69
     decept
    0.66
    这个
    0.63
    这款
    0.63
     aptly
    0.63
    此类
    0.62
     denna
    0.60
     obej
    0.60
     aquesta
    0.60
     intitul
    0.57
    Activations Density 0.155%

    No Known Activations