© 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. 1455
    Prev
    Next
    INDEX
    Explanations

    potential and planned

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

    This neuron highlights qualifying adjectives that modify types of relationships or networks (e.g. “indirect,” “potential”).

    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
     apapun
    0.60
    a
    0.59
    .
    0.59
     każ
    0.57
    Ра
    0.57
     ricorda
    0.57
    ă
    0.55
    ,.
    0.55
    मैं
    0.54
     ću
    0.54
    POSITIVE LOGITS
     proximate
    0.73
    rogens
    0.72
     ఇతర
    0.63
     isoforms
    0.61
     unmodified
    0.61
     combinatorial
    0.61
     त्यानंतर
    0.60
     unrecognized
    0.60
    ogeneous
    0.59
     neighboring
    0.58
    Activations Density 0.070%

    No Known Activations