© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6176
    Prev
    Next
    INDEX
    Explanations

    image cropping/resizing coordinates

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

    The neuron is selectively activating on proper names and branded/named entities (e.g. “UAE,” “Qatar,” “Congo,” “Learning,” “Niari,” 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
    apkan
    0.57
    ての
    0.56
     yakni
    0.55
     лигасы
    0.54
    existent
    0.54
    więks
    0.53
    ]^{
    0.52
    ellikle
    0.51
    年から
    0.51
    劶
    0.51
    POSITIVE LOGITS
    <eos>
    0.64
    Alternatively
    0.55
     FHWA
    0.55
     उत्तराखंड
    0.54
    .
    0.54
     Fou
    0.54
    .$
    0.54
     Uttarakhand
    0.54
     Allora
    0.53
     Figma
    0.52
    Activations Density 0.607%

    No Known Activations