© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 99824
    Prev
    Next
    INDEX
    Explanations

    coaxial cable, nanomaterials, command

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

    The neuron fires on direct second-person address—places where the text speaks to “you” (e.g. “let me show you,” “when you…,” “your…”).

    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
    -1.53
    ๙
    -1.48
     suyo
    -1.47
    weren
    -1.41
    mazda
    -1.41
     kind
    -1.40
    igth
    -1.39
    obras
    -1.38
     ons
    -1.36
    ▾
    -1.36
    POSITIVE LOGITS
    Leírás
    1.83
    2
    1.77
    Кроме
    1.73
     також
    1.55
    Также
    1.54
    5
    1.53
    Склад
    1.51
    Transk
    1.50
    Симпто
    1.47
    Таким
    1.46
    Activations Density 0.025%

    No Known Activations