© 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. 61360
    Prev
    Next
    INDEX
    Explanations

    InteropServices, fractured, Xer, thresh, disjoint

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

    The neuron fires on relatively rare or specialized content words—often multisyllabic technical or descriptive terms (e.g. “InteropServices,” “oblique,” “fractured,” “threshing”).

    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
    archiwizowane
    -3.56
     zoude
    -3.03
    蚨
    -2.94
    豳
    -2.86
    OUTUBE
    -2.78
    瘰
    -2.75
     pokaz
    -2.70
     veicolo
    -2.69
     traffico
    -2.67
    
    -2.67
    POSITIVE LOGITS
    </h5>
    3.27
    ?”
    3.17
     &
    2.91
    getMenuInflater
    2.73
    ]
    2.59
     они
    2.55
     «
    2.50
     ”
    2.44
    x
    2.42
    
    2.33
    Activations Density 0.029%

    No Known Activations