© 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. 10-GEMMASCOPE-RES-131K
    4. 9666
    Prev
    Next
    INDEX
    Explanations

    best friend

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

    The neuron strongly activates on mentions of “friend,” especially when paired with “best,” i.e. the phrase “best friend.”

    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
     a
    -1.57
     if
    -1.55
     simply
    -1.45
    )
    -1.42
     due
    -1.39
    bil
    -1.38
    gister
    -1.38
    tic
    -1.38
    ↵↵↵↵↵
    -1.36
    ">
    -1.34
    POSITIVE LOGITS
     things
    1.79
     میشود
    1.37
     kommende
    1.34
     gelegt
    1.33
     geschlagen
    1.32
    lindo
    1.29
     procesu
    1.27
     schuldig
    1.27
     *
    1.23
     personalizados
    1.21
    Activations Density 0.091%

    No Known Activations