© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10430
    Prev
    Next
    INDEX
    Explanations

    expressing desire or intent

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

    The neuron fires on the modal verb “would.”

    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
    t
    -1.80
    on
    -1.77
    n
    -1.65
    栻
    -1.52
     reverend
    -1.49
    new
    -1.48
    when
    -1.47
     diseñar
    -1.47
    ஂ
    -1.45
    呖
    -1.41
    POSITIVE LOGITS
     espec
    1.46
     particuli
    1.43
     especial
    1.41
    cucina
    1.39
    が欲しい
    1.38
    Denne
    1.33
    thumbsup
    1.33
    drawRect
    1.32
    鸶
    1.30
     hopes
    1.30
    Activations Density 0.039%

    No Known Activations