© 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. 128965
    Prev
    Next
    INDEX
    Explanations

    Christoph and other names

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

    The neuron fires on isolated “named” tokens—proper nouns, technical acronyms, or code identifiers (e.g. class and method names).

    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
    ↵
    -3.48
    時間
    -2.88
    They
    -2.86
     When
    -2.84
    The
    -2.83
    {
    -2.78
        
    -2.69
    c
    -2.59
     \
    -2.58
    很
    -2.50
    POSITIVE LOGITS
    vantaged
    3.92
    
    2.84
    耵
    2.70
     historische
    2.67
    罍
    2.56
    ⩥
    2.56
    硨
    2.55
    笵
    2.55
     donnent
    2.53
    鲠
    2.44
    Activations Density 0.002%

    No Known Activations