© 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. 6197
    Prev
    Next
    INDEX
    Explanations

    predictive maintenance

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

    This neuron detects the subword “pre” (i.e. tokens beginning with the prefix “pre”).

    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
     doorways
    0.60
    上半年
    0.59
     đáy
    0.58
    頪
    0.58
     ആദ്യം
    0.56
    zął
    0.56
    .$\
    0.55
    还在
    0.54
    后再
    0.54
     mellett
    0.54
    POSITIVE LOGITS
     Pre
    1.30
    Pre
    1.28
     PRE
    1.06
    PRE
    1.03
     Preheat
    1.03
    Pré
    1.00
     Pré
    1.00
    pre
    0.99
     प्री
    0.99
    预
    0.99
    Activations Density 0.286%

    No Known Activations