© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 81451
    Prev
    Next
    INDEX
    Explanations

    words starting with Mand

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

    The neuron flags any occurrence of the substring “mand” (as in “Mandriva,” “mandrels,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    unile
    -0.85
    wards
    -0.81
     Disorder
    -0.79
    VarChar
    -0.79
     ramps
    -0.79
    izan
    -0.79
    美好
    -0.78
    ajas
    -0.78
     Clu
    -0.78
    kaner
    -0.75
    POSITIVE LOGITS
    Mand
    1.31
     Mand
    1.14
     MAND
    1.10
    MAND
    0.99
     mand
    0.90
     manda
    0.88
     faca
    0.82
     Mandate
    0.82
     mandato
    0.76
     Mandala
    0.74
    Activations Density 0.017%

    No Known Activations