© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26947
    Prev
    Next
    INDEX
    Explanations

    phosphorylation, separation, IoT, interstitial, carcinoma, World, Scene, career, Press, format

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

    scientific and technical terms in academic or research contexts.

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

    The neuron fires on big, standalone technical headings or key section titles (e.g. “Appendix,” “Chromatin,” “Tyrosine,” “Cheapest IoT Module”) that mark major divisions or titles in scientific or technical documents.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    getreten
    -1.05
     preparando
    -1.05
     kommunen
    -1.04
    竟然是
    -1.04
     unangemessene
    -1.03
     gwia
    -1.02
     nauczy
    -1.02
    podar
    -1.02
     dlouho
    -1.02
    리를
    -1.02
    POSITIVE LOGITS
     prode
    1.04
     stesse
    1.04
     voda
    1.01
    푼
    0.98
    setInitial
    0.97
     versioni
    0.96
     esperienze
    0.96
    Sonido
    0.96
     notizie
    0.96
     maje
    0.96
    Activations Density 0.074%

    No Known Activations