© 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. 94124
    Prev
    Next
    INDEX
    Explanations

    code snippets

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

    The neuron fires on metadata‐style tokens—numbers, version or date codes, identifiers and author/file names, and other punctuation‐heavy or structured “non-content” tokens.

    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
    CountryCode
    -0.74
     照明
    -0.72
    kannten
    -0.71
    ꧂
    -0.69
    ODES
    -0.69
    young
    -0.69
    𝓬
    -0.68
     getCategory
    -0.67
    わけではない
    -0.67
    nodeType
    -0.67
    POSITIVE LOGITS
    スム
    0.72
     Morg
    0.71
    pillar
    0.68
    Sites
    0.68
     aks
    0.66
    뼈
    0.65
    ОВА
    0.65
    semb
    0.64
     filter
    0.63
     الموقع
    0.63
    Activations Density 0.065%

    No Known Activations