© 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. 48077
    Prev
    Next
    INDEX
    Explanations

    verbs followed by objects or results

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

    This neuron fires on third-person singular present-tense verb forms (e.g. “damages,” “takes,” “populates,” “pollutes”).

    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
    -1.48
    那些
    -1.44
    Last
    -1.41
    Again
    -1.41
    problems
    -1.40
    这些
    -1.30
    something
    -1.30
    去年
    -1.28
     also
    -1.28
    hamed
    -1.26
    POSITIVE LOGITS
     this
    1.99
     from
    1.80
     which
    1.79
     before
    1.48
     you
    1.38
     keep
    1.35
     our
    1.34
     such
    1.34
     their
    1.30
     but
    1.29
    Activations Density 0.146%

    No Known Activations