© 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. 40126
    Prev
    Next
    INDEX
    Explanations

    what can be done

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

    This neuron fires on question phrases asking “What can…?” or “Can…?” (e.g. “What can I do…?”, “What can be done…?”, “Can it offer…?”).

    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
    鈊
    -1.92
     土産
    -1.91
     收納
    -1.88
     alip
    -1.87
    ↑↑
    -1.87
    鶲
    -1.87
     vider
    -1.78
    桖
    -1.77
    ♪♪
    -1.76
    　　（
    -1.74
    POSITIVE LOGITS
    ↵↵
    2.13
    5
    1.94
     "
    1.94
    u
    1.93
    0
    1.74
     -
    1.72
     |
    1.69
    守り
    1.66
    g
    1.61
    </h4>
    1.60
    Activations Density 0.026%

    No Known Activations