© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 13468
    Prev
    Next
    INDEX
    Explanations

    when or after state

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

    The neuron strongly activates on words that denote presenting or showcasing concrete results or data (e.g. graphs, insights, highlights).

    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_31_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
    だけでなく
    0.97
    はもちろん
    0.93
    뿐
    0.89
    に加え
    0.88
     மட்டுமல்ல
    0.87
     પરંતુ
    0.82
     поскольку
    0.81
     while
    0.81
     जबकि
    0.79
     insofar
    0.78
    POSITIVE LOGITS
    和
    0.87
     ও
    0.79
     and
    0.76
    和服务
    0.75
    และความ
    0.70
     و
    0.68
    และการ
    0.68
    和社会
    0.68
    และ
    0.66
     وم
    0.66
    Activations Density 0.120%

    No Known Activations