© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 2205
    Prev
    Next
    INDEX
    Explanations

    documents and data formats

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

    This neuron is strongly activated by the period (dot) character as used in code or structured references.

    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
     amiga
    -0.73
    untura
    -0.70
    coro
    -0.69
    UNIC
    -0.68
    𝚜
    -0.67
    ე
    -0.66
    enteuer
    -0.65
     пот
    -0.65
     acom
    -0.64
    tikum
    -0.64
    POSITIVE LOGITS
     مُعرِّف
    0.75
     consens
    0.69
    ही
    0.66
     cached
    0.66
    開店
    0.64
     Pomer
    0.64
    Hib
    0.63
    bori
    0.63
     รัก
    0.61
     ক
    0.60
    Activations Density 0.068%

    No Known Activations