© 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. 22-GEMMASCOPE-RES-131K
    7. 15127
    Prev
    Next
    INDEX
    Explanations

    detecting common phrases

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

    The neuron fires on specialized jargon and technical terminology (e.g. domain‐specific nouns and acronyms).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    </h1>
    -1.51
     on
    -1.47
     you
    -1.43
     
    -1.42
     „
    -1.41
    の声
    -1.34
    だね
    -1.24
     are
    -1.23
    ほうが
    -1.20
     anxiety
    -1.20
    POSITIVE LOGITS
    <bos>
    1.61
    がありますが
    1.54
    ることができます
    1.50
     (~(
    1.48
    があり
    1.48
     모든
    1.45
     '';
    1.43
    そういった
    1.43
    虽说
    1.41
    但这
    1.40
    Activations Density 0.183%

    No Known Activations