© 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. 87246
    Prev
    Next
    INDEX
    Explanations

    vessel walls or junctions

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

    The neuron detects references to containers or ships, specifically the word "vessel" and related terms.

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

    The neuron is chiefly detecting occurrences of the word “vessel.”

    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
    硨
    -3.64
    鐿
    -2.84
    仉
    -2.80
     visse
    -2.78
     soldat
    -2.61
    媜
    -2.52
     kredi
    -2.44
    麃
    -2.44
    缑
    -2.41
     adulte
    -2.39
    POSITIVE LOGITS
    to
    3.19
    2.98
    What
    2.98
    It
    2.83
     This
    2.75
    If
    2.59
     Änderungen
    2.55
    л
    2.47
    Are
    2.45
    This
    2.42
    Activations Density 0.007%

    No Known Activations