© 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. 22-GEMMASCOPE-RES-131K
    7. 39766
    Prev
    Next
    INDEX
    Explanations

    specific technical words

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

    This neuron strongly activates on programming-related tokens—especially identifiers (function or variable names, often containing underscores or CamelCase) and code symbols.

    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
    ﹞
    -0.88
    Література
    -0.84
     kuiten
    -0.83
     enää
    -0.82
    vyš
    -0.79
    ACG
    -0.78
     custodi
    -0.77
    tét
    -0.76
    тія
    -0.76
    passphrase
    -0.75
    POSITIVE LOGITS
     on
    0.82
     None
    0.79
    None
    0.79
     -
    0.79
     assigned
    0.79
    贡
    0.77
     насо
    0.76
     none
    0.75
    Siehe
    0.75
     моменты
    0.75
    Activations Density 0.066%

    No Known Activations