© 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. 24-GEMMASCOPE-2-RES-16K
    7. 6668
    Prev
    Next
    INDEX
    Explanations

    words after common verbs or prepositions

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

    The neuron fires most strongly on the kind of cue‐phrases and verbs used to introduce tips or “how-to” instructions (e.g. “Listed,” “here,” “how,” “to,” “deal,” “making”).

    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_24_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
    $.
    0.80
    $,
    0.74
    마트
    0.71
     foldable
    0.71
    $/
    0.69
     "/
    0.68
    \%$
    0.68
    $',
    0.68
    🎫
    0.68
    lemma
    0.66
    POSITIVE LOGITS
     Май
    0.85
     Haci
    0.85
    capi
    0.81
    cí
    0.80
    േ
    0.79
    cenia
    0.79
     стари
    0.77
    ان
    0.76
     язы
    0.75
     Microsc
    0.75
    Activations Density 0.001%

    No Known Activations