© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15529
    Prev
    Next
    INDEX
    Explanations

    describing environments or states

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

    The neuron strongly activates on isolated, content‐bearing nouns and abstract concept words (e.g. “diversity,” “process,” “factor,” “values,” “service”), i.e. key topic or theme words.

    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
    0.84
     Nhiều
    0.68
    相结合
    0.66
    ,
    0.64
     Архивная
    0.63
     Empirical
    0.62
     Also
    0.62
     *
    0.61
     Commercial
    0.61
     Brief
    0.61
    POSITIVE LOGITS
    which
    1.42
     which
    1.30
     котором
    1.23
     welche
    1.15
     которого
    1.15
     lesquels
    1.13
     которому
    1.11
     которое
    1.10
     которой
    1.09
     lequel
    1.09
    Activations Density 0.306%

    No Known Activations