© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 113293
    Prev
    Next
    INDEX
    Explanations

    Camellia sinensis and {$ code }

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

    code or text segments containing the term "is" followed by certain values or descriptors.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron specifically detects occurrences of the copula “is.”

    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
    あります
    -1.91
    
    -1.86
    </b>
    -1.56
     yapılır
    -1.56
    代わりに
    -1.52
    opheles
    -1.52
     sayesinde
    -1.50
     スイーツ
    -1.50
     anahtar
    -1.49
    śl
    -1.47
    POSITIVE LOGITS
     then
    1.93
     they
    1.70
     bombe
    1.60
     the
    1.56
    nies
    1.44
    \
    1.42
    一个
    1.38
     condominio
    1.36
    −/−
    1.36
     변경
    1.34
    Activations Density 0.002%

    No Known Activations