© 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. 127975
    Prev
    Next
    INDEX
    Explanations

    preventable causes

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

    The neuron fires strongly on closing quotation marks attached to quoted words or phrases (e.g. title names or dialogue) at their ends.

    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
     to
    -2.61
    却被
    -2.45
    rightarrow
    -2.23
     We
    -2.19
     was
    -2.09
    ר
    -2.09
     esteja
    -2.06
     刺繍
    -2.06
    餿
    -2.05
    dhist
    -2.03
    POSITIVE LOGITS
    to
    3.28
    </h3>
    3.25
    with
    3.23
    </i>
    2.88
    時間も
    2.69
    which
    2.64
    Their
    2.50
    4
    2.48
    くらいで
    2.47
    They
    2.42
    Activations Density 0.001%

    No Known Activations