© 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. 84613
    Prev
    Next
    INDEX
    Explanations

    key or important concepts

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

    The neuron activates on words signalling importance or centrality—especially “key” (and sometimes “important”)—highlighting pivotal concepts or steps.

    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
    )
    -1.85
    </i>
    -1.81
    ль
    -1.53
    }
    -1.52
     ค
    -1.52
    喾
    -1.49
    ;
    -1.48
     This
    -1.42
    蟊
    -1.41
     Recommended
    -1.40
    POSITIVE LOGITS
     they
    2.17
     terutama
    2.08
     also
    1.89
    also
    1.55
    尤其
    1.54
    尤其是
    1.53
     должен
    1.49
    两人的
    1.49
    ！）
    1.48
    レンタ
    1.48
    Activations Density 0.007%

    No Known Activations