© 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. 34-GEMMASCOPE-RES-131K
    7. 106907
    Prev
    Next
    INDEX
    Explanations

    Hot Country

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

    The neuron strongly activates on copular or linking verbs (forms of “to be” such as is, was, are, as, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Jana
    -0.81
    秀
    -0.65
    thoff
    -0.64
     θε
    -0.63
    déric
    -0.63
    TCC
    -0.61
    riez
    -0.60
    qid
    -0.60
     knees
    -0.59
     hips
    -0.58
    POSITIVE LOGITS
    epoca
    0.73
    
    0.72
     Вікіпе
    0.71
    жное
    0.69
     エク
    0.68
    ruik
    0.68
    amous
    0.68
    uchs
    0.68
    CONTROL
    0.67
    HEAST
    0.67
    Activations Density 0.141%

    No Known Activations