© 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. 6569
    Prev
    Next
    INDEX
    Explanations

    identifies exact balance

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

    The neuron fires most strongly on words and tokens that refer to downloadable digital resources or file downloads (e.g. “download,” “PDF,” “EBOOK,” “Gestalt Therapist” as a downloadable listing, etc.).

    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
     وهو
    0.78
     tetapi
    0.75
    첨
    0.74
     وهي
    0.73
     التي
    0.72
     وعلى
    0.71
    Kemudian
    0.70
     যখন
    0.70
     仕
    0.68
     haline
    0.67
    POSITIVE LOGITS
    ামুটি
    0.87
    り
    0.86
     eukaryotes
    0.84
    lung
    0.84
    dni
    0.82
     ECL
    0.80
    straat
    0.80
    ৯০
    0.80
    thirds
    0.78
     Counc
    0.78
    Activations Density 0.012%

    No Known Activations