© 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. 1982
    Prev
    Next
    INDEX
    Explanations

    Final Answer: The final answer is

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

    The neuron fires on words and phrases used to point readers toward related images or other content (e.g. “similar,” “related,” “pictures,” “scroll,” “collection”).

    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
     authent
    0.59
    <h2>
    0.59
    aske
    0.58
    .$
    0.56
     superf
    0.55
    .’”
    0.55
     fi
    0.54
     gent
    0.53
    ,’”
    0.53
     account
    0.52
    POSITIVE LOGITS
    </h4>
    0.84
    youtube
    0.67
    Youtube
    0.65
     Youtube
    0.64
     এরূপ
    0.64
     youtube
    0.62
    consulta
    0.61
    спорт
    0.58
     प्रश्नावली
    0.58
    骼
    0.58
    Activations Density 0.009%

    No Known Activations