© 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. 81096
    Prev
    Next
    INDEX
    Explanations

    virulent, cyclogenesis, rAAV, atherogenesis, irinotecan

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

    technical/scientific terms, particularly chemical compounds, medications, and biological concepts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on domain‐specific uppercase abbreviations and acronyms (e.g. technical/library names like JNI or chemical/biological codes like AAV, ACV, cyclo-, vir, ther, etc.).

    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
     anything
    -1.12
    ườ
    -0.83
    dây
    -0.83
     Aufmerksamkeit
    -0.82
    ئے
    -0.82
     gewöhnlich
    -0.79
    کری
    -0.79
    首先
    -0.79
    itamente
    -0.78
    etheless
    -0.77
    POSITIVE LOGITS
     solicitation
    0.81
    馴染
    0.76
    *}{
    0.73
     Excellence
    0.73
    助力
    0.72
     Stunden
    0.71
    诹
    0.71
    chè
    0.70
     isSuccess
    0.70
    ں
    0.70
    Activations Density 0.196%

    No Known Activations