© 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. 57754
    Prev
    Next
    INDEX
    Explanations

    you are looking for

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

    The neuron flags direct second-person address and call-to-action phrases—e.g. “If you are…,” “Are you looking for…,” “Whether you need…”.

    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
     tende
    -0.91
     vieles
    -0.90
     הרבה
    -0.86
    ίνεται
    -0.86
     掃除
    -0.84
    ことができる
    -0.84
     volunteering
    -0.83
     conduce
    -0.82
     adept
    -0.82
    Kosten
    -0.81
    POSITIVE LOGITS
     require
    2.02
     need
    1.91
     requires
    1.67
    需要
    1.64
     نیاز
    1.62
     requiring
    1.59
     requirement
    1.59
     Require
    1.56
     cần
    1.53
    Require
    1.52
    Activations Density 0.054%

    No Known Activations