© 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. 34-GEMMASCOPE-RES-131K
    7. 2139
    Prev
    Next
    INDEX
    Explanations

    This project

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

    The neuron fires on generic academic framing and meta‐discourse phrases—especially funding or acknowledgment language (e.g. “This project,” “partially funded,” “This research,” “work described in this paper,” “thank”).

    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
    ktan
    -0.74
    Savage
    -0.70
    portada
    -0.70
     UIApplication
    -0.68
     onderdeel
    -0.68
     Massen
    -0.67
    商人
    -0.67
     Ça
    -0.67
     integrados
    -0.67
    ૅ
    -0.67
    POSITIVE LOGITS
    BUILD
    0.71
    assas
    0.71
     sabes
    0.70
     расши
    0.69
    обо
    0.68
     упо
    0.67
     asiático
    0.66
    ⁄
    0.66
    孩子的
    0.66
    build
    0.66
    Activations Density 0.061%

    No Known Activations