© 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. 73537
    Prev
    Next
    INDEX
    Explanations

    titles of works

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

    The neuron activates on multiword, capitalized titles of creative works (films, books, graphic novels, 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_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
     СИ
    -1.50
     ></
    -1.45
     and
    -1.45
     ТАК
    -1.41
     ГО
    -1.39
     cloths
    -1.36
     pierwszym
    -1.34
    voye
    -1.34
     ЛИ
    -1.33
     ОБ
    -1.32
    POSITIVE LOGITS
     two
    1.68
    !
    1.50
     Säure
    1.46
     data
    1.45
    </b>
    1.44
     six
    1.43
     widerrufen
    1.41
     prodot
    1.38
     pido
    1.38
     nuevas
    1.37
    Activations Density 0.090%

    No Known Activations