© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 67946
    Prev
    Next
    INDEX
    Explanations

    code package suffixes

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

    The neuron fires on tokens that are parts of dotted package or import/module paths (i.e. segments of fully qualified names in code).

    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
     on
    -2.47
     for
    -2.33
     at
    -2.17
     this
    -2.14
     from
    -2.06
     even
    -2.05
     in
    -2.02
     that
    -2.00
     with
    -1.96
     FOR
    -1.95
    POSITIVE LOGITS
    があり
    1.84
     développer
    1.80
     publier
    1.66
     faciliter
    1.59
     diriger
    1.59
    possibly
    1.56
    这让
    1.55
    
    1.55
     essas
    1.55
    currently
    1.55
    Activations Density 0.031%

    No Known Activations