© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 41938
    Prev
    Next
    INDEX
    Explanations

    generally, usually, currently, actually

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

    The neuron activates on adverbs—particularly those ending in “-ly,” like generally, usually, and currently.

    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
     there
    -1.71
     vagy
    -1.70
    蟜
    -1.66
    THEY
    -1.64
     Conſ
    -1.55
     either
    -1.54
    还不错
    -1.53
     wither
    -1.51
    
    -1.49
    either
    -1.46
    POSITIVE LOGITS
    When
    2.05
    This
    1.86
    Because
    1.81
    Even
    1.81
    </em>
    1.73
    when
    1.73
    During
    1.70
    These
    1.68
    。”
    1.66
    While
    1.64
    Activations Density 0.020%

    No Known Activations