© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 81671
    Prev
    Next
    INDEX
    Explanations

    describe(" followed by text

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

    The neuron fires strongly on programming identifiers (library/class names, function names, section‐header keywords, etc.) rather than ordinary English 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
    anneau
    -0.76
    teilen
    -0.73
    msubsup
    -0.73
    spielen
    -0.71
    sprzed
    -0.70
    suchen
    -0.70
    powiedz
    -0.69
    
    -0.69
     ‏
    -0.69
     Că
    -0.69
    POSITIVE LOGITS
     subtitle
    0.76
     pamphlet
    0.69
     Wich
    0.69
    ᱾
    0.68
     निम्
    0.67
     nenhuma
    0.66
    sld
    0.65
    ณี
    0.65
     subtitles
    0.64
    體驗
    0.64
    Activations Density 0.053%

    No Known Activations