© 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. 128147
    Prev
    Next
    INDEX
    Explanations

    nitrous oxide, criss-cross, Marian names

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

    The neuron identifies proper names and chemical compounds like "nitrous oxide" and "Marianna".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    s
    -3.25
    ↵
    -3.20
    n
    -3.08
    あります
    -2.61
     był
    -2.53
    lgari
    -2.42
    o
    -2.42
    D
    -2.39
    a
    -2.38
    喽
    -2.38
    POSITIVE LOGITS
     sprüche
    2.58
    ﹶ
    2.56
     помогают
    2.48
    仉
    2.47
    轳
    2.47
    ໑
    2.42
     Appetit
    2.41
     ferrovia
    2.34
     selben
    2.34
    ING
    2.28
    Activations Density 0.024%

    No Known Activations