© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26363
    Prev
    Next
    INDEX
    Explanations

    continue reading

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

    The neuron fires on structural markers of truncated or continued content (e.g. “Read More,” “…,” and the numeric preview lengths).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     it
    -1.05
    マヨ
    -0.96
    寝
    -0.92
     anything
    -0.91
     Verder
    -0.91
    ↵
    -0.90
     バッジ
    -0.81
     dikatakan
    -0.81
     these
    -0.79
    boven
    -0.77
    POSITIVE LOGITS
    酐
    0.85
    maids
    0.85
     ilha
    0.85
    几次
    0.85
    LogFactory
    0.84
     الفلسط
    0.84
    ous
    0.84
    ła
    0.84
     кү
    0.83
    Addiction
    0.83
    Activations Density 0.008%

    No Known Activations