© 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. 38965
    Prev
    Next
    INDEX
    Explanations

    verbs followed by prepositions/articles

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

    The neuron fires on “-ing” words that signal someone’s involvement, desire, or action (e.g. interested, planning, seeking, undergoing, giving).

    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
     also
    -1.77
    ,
    -1.60
     cambiado
    -1.53
     vuelto
    -1.47
     about
    -1.39
     sabido
    -1.39
    ждает
    -1.37
     cojín
    -1.35
    penuhi
    -1.33
    
    -1.33
    POSITIVE LOGITS
     who
    1.88
     their
    1.85
     الذين
    1.78
    their
    1.66
     to
    1.59
     themselves
    1.56
    錞
    1.47
    谁
    1.41
     ktorí
    1.39
     majú
    1.38
    Activations Density 0.137%

    No Known Activations