© 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. 13032
    Prev
    Next
    INDEX
    Explanations

    refers to

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

    The neuron fires on definitional phrasing, particularly instances of “refers to” introducing a term’s meaning.

    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
    ল
    0.81
     st
    0.80
     cl
    0.77
     d
    0.75
    0.74
    分为
    0.73
    ↵
    0.71
     array
    0.71
     definitions
    0.71
     an
    0.70
    POSITIVE LOGITS
     Также
    0.93
    無論
    0.92
     Bahkan
    0.91
     Даже
    0.91
    Escolhido
    0.91
     любой
    0.89
     любую
    0.89
    quando
    0.89
     Mereka
    0.88
    ,_-
    0.87
    Activations Density 0.005%

    No Known Activations