© 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. 84179
    Prev
    Next
    INDEX
    Explanations

    respect to or relation to

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

    The neuron detects legal‐style reference phrases that express relation or scope—e.g. “with respect to,” “relating to,” and “relates to.”

    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
    方法は
    -0.86
     or
    -0.83
    amkeit
    -0.81
     pokrač
    -0.81
    kunci
    -0.80
    Kama
    -0.80
    :✨
    -0.79
     благодаря
    -0.79
     entendi
    -0.78
     réalité
    -0.78
    POSITIVE LOGITS
     matters
    1.30
     cuyo
    1.07
     هر
    1.03
    any
    1.00
    こと
    0.98
     events
    0.97
     cuyas
    0.94
     such
    0.93
    Anything
    0.92
    ことに
    0.91
    Activations Density 0.018%

    No Known Activations