© 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. 129053
    Prev
    Next
    INDEX
    Explanations

    greetings and well wishes

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

    The neuron fires on line‐start markers in multi‐message threads—blank separators, greetings (“Hello,” “Hi,” “I”), headers (names like “Kay”), and similar tokens that signal the beginning of a new post or email segment.

    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
    }\
    -1.55
    鉿
    -1.55
    珦
    -1.51
     the
    -1.45
    ꦼ
    -1.42
    *,
    -1.41
    袿
    -1.36
    衚
    -1.35
    镠
    -1.34
    亿元
    -1.33
    POSITIVE LOGITS
    🫃
    1.64
    姈
    1.64
    ,
    1.55
    vetrain
    1.55
     其
    1.54
     Also
    1.54
     Either
    1.51
     Soporte
    1.49
    issant
    1.48
    ᥬ
    1.48
    Activations Density 0.023%

    No Known Activations