© 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. 102485
    Prev
    Next
    INDEX
    Explanations

    wishing to do something

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

    The neuron selectively activates on the modal verb “could.”

    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
     materiál
    -1.61
     úteis
    -1.56
     potrebné
    -1.52
    ৄ
    -1.50
    ࿊
    -1.49
     Fähigkeit
    -1.48
    )$
    -1.46
    máy
    -1.45
     Begleitung
    -1.45
     vhodné
    -1.44
    POSITIVE LOGITS
    </b>
    1.97
    C
    1.59
     vince
    1.44
     secr
    1.43
    uk
    1.42
    t
    1.42
    se
    1.40
    за
    1.39
    PS
    1.39
    errit
    1.38
    Activations Density 0.007%

    No Known Activations