© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9334
    Prev
    Next
    INDEX
    Explanations

    URLs, downloads, cuda, wiki

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

    The neuron detects numerical answer‐score values (e.g. the floating‐point numbers following “Answer:” or “BEST ANSWER:”).

    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
     scrapbook
    0.66
    🇴
    0.59
    illy
    0.59
    hanga
    0.57
    ুবাদ
    0.55
    ৃত্ব
    0.55
     notebook
    0.54
    decrement
    0.54
     پاورپوینت
    0.54
    ਿੱ
    0.54
    POSITIVE LOGITS
    <start_of_image>
    0.74
    连忙
    0.63
    Just
    0.61
    แนะนำ
    0.60
    担任
    0.59
    一番
    0.57
    All
    0.56
    Table
    0.56
    通常
    0.55
    Cic
    0.55
    Activations Density 0.000%

    No Known Activations