© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 105593
    Prev
    Next
    INDEX
    Explanations

    column headers or dataframe fields

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

    The neuron consistently activates on tokens that look like table or dataframe column headers (e.g. field names) in code-oriented examples.

    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
    simply
    -1.70
     فقط
    -1.63
    
    -1.59
    ⟣
    -1.58
    ⎔
    -1.56
    锨
    -1.55
     gouverneur
    -1.54
     tête
    -1.53
     interessant
    -1.52
     majest
    -1.49
    POSITIVE LOGITS
     from
    1.84
     people
    1.76
     allows
    1.67
     makes
    1.66
     with
    1.60
     on
    1.60
     for
    1.55
     provide
    1.54
     brings
    1.52
     set
    1.52
    Activations Density 0.024%

    No Known Activations