© 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. 34-GEMMASCOPE-RES-131K
    7. 30956
    Prev
    Next
    INDEX
    Explanations

    inflammatory disease

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

    license and copyright text in programming code and legal documentation.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly responds to non-ASCII characters (e.g. Hebrew letters and diacritics) rather than ordinary Latin text.

    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
    Alexandre
    -0.92
    Compras
    -0.84
    culous
    -0.83
    Efe
    -0.82
    valuable
    -0.82
    monument
    -0.82
    Imgur
    -0.82
    southern
    -0.75
     keb
    -0.75
     perfons
    -0.73
    POSITIVE LOGITS
     egli
    0.92
    ס
    0.92
    忌
    0.90
     cerevisiae
    0.87
    Father
    0.86
    ঙ
    0.86
    CTYPE
    0.84
     spake
    0.83
    Finish
    0.82
    (""))
    0.82
    Activations Density 0.008%

    No Known Activations