© 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. 10-GEMMASCOPE-RES-131K
    7. 45213
    Prev
    Next
    INDEX
    Explanations

    Overview followed by context

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

    The neuron strongly activates on standalone section headings or document “Overview”-style titles.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     to
    -1.82
     ผม
    -1.42
    旆
    -1.42
    igia
    -1.42
     owing
    -1.40
    śród
    -1.38
     christlichen
    -1.36
     který
    -1.33
     Einfluß
    -1.32
     That
    -1.30
    POSITIVE LOGITS
    ogenicity
    1.52
    涧
    1.47
    \
    1.46
    itability
    1.44
    时期
    1.44
    natured
    1.41
    tiness
    1.40
     blancs
    1.38
    burner
    1.33
    asters
    1.32
    Activations Density 0.030%

    No Known Activations