© 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. 6385
    Prev
    Next
    INDEX
    Explanations

    contact a cushion

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

    The neuron is highly activated by decimal‐formatted numbers (floating‐point numeric tokens) typically found in citation metadata.

    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
     UNNEEDED
    0.85
     вижу
    0.85
    役に
    0.82
     capazes
    0.81
    е
    0.80
     глаза
    0.79
     রাখ
    0.79
    гов
    0.78
    隐含
    0.78
    আমাকে
    0.78
    POSITIVE LOGITS
    AT
    0.89
    U
    0.88
    u
    0.86
    IC
    0.84
    H
    0.83
    ET
    0.82
    Unit
    0.82
    EP
    0.80
    AC
    0.79
    EM
    0.77
    Activations Density 0.001%

    No Known Activations