© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43631
    Prev
    Next
    INDEX
    Explanations

    structured data delimiters

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

    This neuron detects identifiers used as field, column, or key names (e.g. struct member names, DataFrame column headers, JSON tags).

    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
    埽
    -2.72
    秾
    -2.70
     humeur
    -2.69
     articoli
    -2.64
     stalwart
    -2.61
     pittura
    -2.59
    ⎚
    -2.58
    孭
    -2.55
    ᕦ
    -2.55
    箜
    -2.55
    POSITIVE LOGITS
    <bos>
    3.97
    ↵
    3.34
     After
    3.11
    ftrag
    2.66
    ～。
    2.59
    vantaged
    2.48
     Apparently
    2.45
    </h3>
    2.44
     For
    2.44
    ).
    2.42
    Activations Density 0.083%

    No Known Activations