© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 76913
    Prev
    Next
    INDEX
    Explanations

    I should get

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

    This neuron fires on parts of the text that are asking a question or stating a request—in particular on interrogative phrasing and first‐person request words.

    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
     a
    -1.99
    larını
    -1.23
     certain
    -1.20
     an
    -1.17
     one
    -1.15
     Beamten
    -1.13
     preuß
    -1.10
    的所有
    -1.06
     Ausführungen
    -1.05
    
    -1.05
    POSITIVE LOGITS
     the
    5.16
     The
    1.45
    ˃
    1.16
    The
    1.13
     הה
    1.11
    ✘
    1.09
     cherchés
    1.07
    ˂
    1.06
     jaunes
    1.05
    咔
    1.05
    Activations Density 0.245%

    No Known Activations