© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 7504
    Prev
    Next
    INDEX
    Explanations

    terms and measurements

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

    The neuron strongly activates on words that signal credibility or endorsement—terms like “English,” “certified,” “analysts’ ratings,” and “reliability” that convey authority or quality assurance.

    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
     denounce
    0.91
    funciones
    0.87
    ificaciones
    0.86
     ediciones
    0.83
    栻
    0.82
     difusión
    0.79
    とその
    0.77
    entityManager
    0.75
     caminar
    0.74
    が必要
    0.73
    POSITIVE LOGITS
    ار
    0.82
    jamin
    0.77
     BBM
    0.77
    三個
    0.76
    ञ
    0.76
    زء
    0.75
     Four
    0.74
    ब
    0.74
     meisten
    0.71
    雯
    0.70
    Activations Density 0.001%

    No Known Activations