© 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. 22-GEMMASCOPE-RES-131K
    7. 39234
    Prev
    Next
    INDEX
    Explanations

    Javascript variables and code

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

    This neuron fires strongly on programming identifiers and other uncommon, domain‐specific tokens (e.g. variable names, type names, annotation names) rather than on 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_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
     живот
    -0.99
    ulpt
    -0.96
    omial
    -0.94
    cobacterium
    -0.92
    erne
    -0.90
     ialah
    -0.89
    ,"%
    -0.88
    ૮
    -0.88
     mV
    -0.86
    Publica
    -0.85
    POSITIVE LOGITS
    Swagger
    0.91
    他们的
    0.84
     gabinete
    0.83
     these
    0.82
    energie
    0.80
    inier
    0.79
    SetId
    0.77
    MERCE
    0.77
     energetic
    0.75
    limite
    0.75
    Activations Density 0.061%

    No Known Activations