© 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. 22-GEMMASCOPE-RES-131K
    7. 70867
    Prev
    Next
    INDEX
    Explanations

    wound repair and healing

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

    The neuron detects mentions of wounds and wound healing in medical or scientific contexts.

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

    The neuron activates on mentions of the word “wound.”

    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
     handlar
    -2.55
    archiwizowane
    -2.50
     dezelve
    -2.47
     }))
    -2.44
    的時候
    -2.44
    "))
    -2.42
     ]
    -2.38
    Teknik
    -2.36
    regierung
    -2.33
     ถ้า
    -2.33
    POSITIVE LOGITS
     be
    2.58
     
    2.48
     complemented
    2.47
    '
    2.45
    er
    2.41
    AN
    2.34
    In
    2.34
     

    2.30
    AT
    2.20
    <b>
    2.16
    Activations Density 0.020%

    No Known Activations