© 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. 39641
    Prev
    Next
    INDEX
    Explanations

    metadata tags like author and date

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

    The neuron fires on documentation‐style metadata labels (e.g. “Title:”, “Description:”, “@date”, “Time:”, “Author:”, XML/HTML doc tags) and their separators.

    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
    CommonModule
    -1.52
    入る
    -1.48
     we
    -1.47
     it
    -1.41
    台の
    -1.41
     I
    -1.41
    完成です
    -1.40
     welchem
    -1.37
    又在
    -1.33
    laß
    -1.28
    POSITIVE LOGITS
    </strong>
    1.83
    ]
    1.73
    s
    1.59
    Not
    1.55
    You
    1.53
    Going
    1.53
     
    1.52
    Those
    1.48
    Are
    1.47
    </em>
    1.46
    Activations Density 0.012%

    No Known Activations