© 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. 12628
    Prev
    Next
    INDEX
    Explanations

    write `std::` or clarify

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

    The neuron is responding to rare or uncommon words (i.e. low‐frequency tokens) in the text.

    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
    e
    0.75
    xa
    0.72
    HeaderParams
    0.71
     thịt
    0.71
     Empresa
    0.71
    ॰ऍ
    0.70
     cobbl
    0.70
     şirk
    0.69
    eos
    0.69
    didn
    0.68
    POSITIVE LOGITS
    历史
    0.75
    防治
    0.74
     си
    0.74
     freeze
    0.73
    kz
    0.73
    ں
    0.73
    详
    0.73
    erset
    0.71
     getConnection
    0.71
     getState
    0.71
    Activations Density 0.000%

    No Known Activations