© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 1827
    Prev
    Next
    INDEX
    Explanations

    set up or direct others

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

    The neuron is looking for numerical tokens or figures (i.e. sequences of digits and numeric measurements).

    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
    injection
    0.86
    m
    0.83
    o
    0.82
    in
    0.78
    ta
    0.71
    aden
    0.68
     lequel
    0.68
     entgegen
    0.68
    మ
    0.68
    information
    0.68
    POSITIVE LOGITS
     cientos
    1.02
     tienes
    0.93
     cuidadosamente
    0.91
     hablamos
    0.89
     грошы
    0.89
     hojas
    0.88
     tengas
    0.88
     vimos
    0.88
     aclar
    0.88
     decenas
    0.87
    Activations Density 0.000%

    No Known Activations