© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113376
    Prev
    Next
    INDEX
    Explanations

    discussing data formats

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

    the word "format" in technical or programming contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The main thing this neuron does is detect occurrences of the word “format.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     auffi
    -3.09
    囖
    -2.80
    ᆞ
    -2.78
    孭
    -2.78
    ,/
    -2.73
     GOtt
    -2.70
    
    -2.67
     Monfieur
    -2.63
     nahil
    -2.58
    齜
    -2.58
    POSITIVE LOGITS
    In
    3.73
    ”
    3.19
    ”，
    2.91
    To
    2.88
    No
    2.64
    For
    2.59
    !”
    2.59
    So
    2.56
    You
    2.52
    p
    2.52
    Activations Density 0.021%

    No Known Activations