© 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. 128286
    Prev
    Next
    INDEX
    Explanations

    age-related / associated

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

    The neuron is highly sensitive to the word “related” (especially in hyphenated constructs like “age‐related”).

    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
     hijau
    -1.52
    踯
    -1.47
    alties
    -1.46
    蓠
    -1.46
    骉
    -1.44
    aneity
    -1.44
     muda
    -1.42
     bym
    -1.40
     本
    -1.39
     JpaRepository
    -1.39
    POSITIVE LOGITS
      
    1.73
     from
    1.59
     now
    1.47
     their
    1.42
     artık
    1.39
    什么样的
    1.39
    </h2>
    1.38
     don
    1.38
    因为它
    1.36
    ของ
    1.32
    Activations Density 0.219%

    No Known Activations