© 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. 8654
    Prev
    Next
    INDEX
    Explanations

    drug optimization, export declaration, predicted mean

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

    The neuron is keyed to verbs (and verb‐like tokens) that introduce locational or movement information in a report (e.g. “located,” “via,” “hunting”).

    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
    Đáp
    0.80
     slike
    0.77
    ningar
    0.73
     考え
    0.73
    OKR
    0.73
    Литература
    0.73
     Surveys
    0.72
    侗
    0.72
    ﺯ
    0.71
     沒有
    0.70
    POSITIVE LOGITS
    iglich
    0.77
    openide
    0.74
    ک
    0.72
    вок
    0.70
    ಾಗಲೇ
    0.70
    под
    0.70
    uah
    0.70
    поль
    0.69
    edict
    0.69
    ка
    0.68
    Activations Density 0.001%

    No Known Activations