© 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. 3669
    Prev
    Next
    INDEX
    Explanations

    temporal-spatial features

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

    The neuron strongly activates on the word “download” (and to a lesser extent “budgeting”), i.e. it’s detecting references to downloading or resource‐management actions.

    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
     org
    0.72
    開店
    0.71
     国内
    0.71
     Comune
    0.69
    ekar
    0.68
    女
    0.68
    部分
    0.67
    hews
    0.67
     فروش
    0.66
     COMO
    0.65
    POSITIVE LOGITS
     radeon
    0.80
     अंका
    0.80
    ALD
    0.76
     trigonometric
    0.75
    Disconnected
    0.75
     encirc
    0.73
    Ͻ
    0.71
    Abandon
    0.70
    addPreferredGap
    0.70
    Abandoned
    0.68
    Activations Density 0.005%

    No Known Activations