© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10493
    Prev
    Next
    INDEX
    Explanations

    whether something is a good fit

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

    The neuron fires on words conveying evaluations or judgments (e.g. “better,” “perfect,” “desirable,” “qualified,” “good,” “makes sense”).

    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
     research
    0.64
    理论
    0.55
    理論
    0.55
    誉
    0.52
     Research
    0.52
    <0x0D>
    0.52
    <strong>
    0.51
     이론
    0.51
    證
    0.51
    ここに
    0.50
    POSITIVE LOGITS
     ឬ
    0.82
     вариант
    0.80
    )،
    0.79
    )।
    0.78
     প্রমুখ
    0.76
    .),
    0.74
     💪
    0.74
    kanie
    0.73
     প্রভৃতি
    0.70
     kapsamında
    0.70
    Activations Density 0.003%

    No Known Activations