© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 12362
    Prev
    Next
    INDEX
    Explanations

    words and related concepts

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

    The neuron activates on numeric tokens with decimal points—that is, floating-point quantities (often prices, measurement values, or other precise figures).

    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
    ี
    0.75
     Lactobacillus
    0.74
     zucchero
    0.71
     thổ
    0.71
     охла
    0.70
     Aloe
    0.69
     Obamacare
    0.67
    </h3>
    0.67
     сахар
    0.67
     甘
    0.67
    POSITIVE LOGITS
    ঘরে
    0.80
     nodes
    0.80
     அனைத்து
    0.78
    ਨੀ
    0.77
     মুহুর্তে
    0.74
     isSignedIn
    0.74
    attup
    0.73
     molecules
    0.72
    nodes
    0.71
    บริเวณ
    0.71
    Activations Density 0.623%

    No Known Activations