© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 73410
    Prev
    Next
    INDEX
    Explanations

    The focus

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

    This neuron detects occurrences of the phrase “focus” (as in “The focus is on…” or “The focus was on…”).

    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
    </i>
    -2.50
     bienvenida
    -1.73
     muerta
    -1.72
     белая
    -1.65
    
    -1.64
    ❲
    -1.63
     Consumo
    -1.60
    </h1>
    -1.60
     farba
    -1.59
     során
    -1.59
    POSITIVE LOGITS
     of
    2.45
     focus
    1.89
    ではでは
    1.85
    </h2>
    1.72
    c
    1.70
     is
    1.69
    to
    1.68
     then
    1.66
    到现在
    1.60
    </h4>
    1.57
    Activations Density 0.014%

    No Known Activations