© 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. 43829
    Prev
    Next
    INDEX
    Explanations

    can + verb sequences

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

    The neuron activates on modal auxiliary verbs like “can” or “could” (often with “be” or another verb) indicating capability or possibility.

    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
    珦
    -1.93
    ",
    -1.54
    </i>
    -1.46
    图片
    -1.45
    Не
    -1.41
    9
    -1.41
    现在的
    -1.41
    来年
    -1.41
    boucles
    -1.39
    齪
    -1.39
    POSITIVE LOGITS
    er
    1.71
     Они
    1.69
    ato
    1.65
     WITH
    1.65
    with
    1.61
    みたいに
    1.59
    ule
    1.55
     ”
    1.49
    }/*
    1.47
    roughts
    1.47
    Activations Density 0.107%

    No Known Activations