© 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. 102689
    Prev
    Next
    INDEX
    Explanations

    French text and technical terms

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

    the letter "l" (lowercase L) following an apostrophe in French text.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron consistently responds to the single-character elided article (e.g. “l’”) used in French and Italian.

    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
    fordshire
    -2.81
    quelize
    -2.67
     ſte
    -2.58
    ztes
    -2.56
    ,’”
    -2.53
    ᨁ
    -2.47
    咝
    -2.47
    izoen
    -2.47
    逦
    -2.45
    ”，
    -2.42
    POSITIVE LOGITS
    '
    4.84
    >
    3.42
    \
    3.20
     '
    3.16
     This
    2.84
    ]
    2.84
    (
    2.72
    *
    2.48
     very
    2.42
    {
    2.41
    Activations Density 0.002%

    No Known Activations