© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 128716
    Prev
    Next
    INDEX
    Explanations

    cardiac and other anomalies

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

    words or phrases related to anomalies or abnormalities, particularly in scientific or medical contexts.

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

    The neuron fires on words built around the root “anomal–” (e.g. anomaly, anomalies, anomalism), effectively spotting mentions of anomalies.

    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
    
    -2.72
     brazen
    -2.72
    俛
    -2.72
     في
    -2.69
    ⤿
    -2.63
    镲
    -2.63
    泚
    -2.56
    瘆
    -2.52
     мероприятия
    -2.52
     stalwart
    -2.44
    POSITIVE LOGITS
    ↵↵
    2.88
    ↵
    2.86
    {
    2.47
    …..
    2.25
    パーティー
    2.19
    .
    2.19
    v
    2.17
    お客
    2.14
     meticulously
    2.14
     every
    2.09
    Activations Density 0.006%

    No Known Activations