© 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. 10-GEMMASCOPE-RES-131K
    7. 105723
    Prev
    Next
    INDEX
    Explanations

    eye parts and medical conditions

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

    terms related to the eye and pupil anatomy, especially in medical or technical contexts.

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

    The neuron strongly activates on ophthalmology‐related terms, especially those naming eye structures (e.g. pupil/pupillary, retina, uveal).

    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
     was
    -1.70
     against
    -1.50
     they
    -1.48
    ";
    -1.48
     said
    -1.43
     wasn
    -1.40
    “
    -1.38
    瓩
    -1.38
     eigenen
    -1.38
     together
    -1.37
    POSITIVE LOGITS
     ofrecemos
    1.52
    Periodo
    1.43
    Takeaways
    1.40
    Jugador
    1.38
    ⦊
    1.37
    Familia
    1.36
     Kalifor
    1.36
     улыб
    1.35
     mélange
    1.35
    Obtain
    1.34
    Activations Density 0.035%

    No Known Activations