© 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. 127844
    Prev
    Next
    INDEX
    Explanations

    brand news league

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

    The neuron activates on components of multi-word or hyphenated proper names and technical terms—i.e. it flags parts of compound nouns, acronyms, and specialized labels.

    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
     of
    -3.41
     when
    -2.02
     as
    -1.69
     for
    -1.63
     such
    -1.54
     they
    -1.49
     during
    -1.46
    蒟
    -1.45
     like
    -1.45
     from
    -1.44
    POSITIVE LOGITS
     couvrir
    1.67
     travailler
    1.63
    worldwide
    1.54
     pratiquer
    1.54
    extremely
    1.54
     livrer
    1.54
    possibly
    1.48
    
    1.45
    whelming
    1.44
    instead
    1.43
    Activations Density 0.038%

    No Known Activations