© 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. 59552
    Prev
    Next
    INDEX
    Explanations

    event type

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

    This neuron fires on the punctuation and keywords that make up URLs or file‐path references (e.g. “/”, “.py”, “?”, “=”, “hl”, “Request”), so it’s looking for URL or code‐file path tokens.

    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
    Boas
    -0.79
    ucción
    -0.77
    正解
    -0.77
     Infe
    -0.74
     borse
    -0.74
    
    -0.74
     중요
    -0.73
    monument
    -0.73
     perceptive
    -0.73
     등록
    -0.72
    POSITIVE LOGITS
     organizador
    0.82
    StylePriority
    0.80
    إذا
    0.79
    Calla
    0.79
     FACTOR
    0.78
    SON
    0.77
     workers
    0.76
    øse
    0.76
    شكرا
    0.75
     class
    0.75
    Activations Density 0.002%

    No Known Activations