© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 36300
    Prev
    Next
    INDEX
    Explanations

    import statements

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

    The neuron specializes in spotting major document headings or titles (e.g. section names, \title fields, “Introduction” headers).

    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
    が
    -2.66
    を
    -2.22
    es
    -1.85
    など
    -1.82
    かは
    -1.80
    $
    -1.78
    f
    -1.75
    r
    -1.74
    ちなみに
    -1.70
    ฉ
    -1.69
    POSITIVE LOGITS
     burbujas
    2.27
     kontaktu
    2.16
     ganancias
    2.09
     pesados
    2.05
     huellas
    2.03
     aguj
    2.02
     and
    2.02
    Groet
    2.00
     pieles
    1.94
     vlasy
    1.91
    Activations Density 0.000%

    No Known Activations