© 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. 11763
    Prev
    Next
    INDEX
    Explanations

    names and surnames

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

    The neuron strongly activates on capitalized name fragments and other title‐case tokens, i.e. parts of proper names and organizations.

    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
    込めて
    -1.73
     d
    -1.69
    ቍ
    -1.68
    ){
    -1.65
                        
    -1.64
     n
    -1.63
    erráneo
    -1.63
    These
    -1.60
    クター
    -1.56
    衣服
    -1.55
    POSITIVE LOGITS
     the
    1.81
     prachtig
    1.75
    ㈲
    1.73
    嗷
    1.68
     prachtige
    1.66
    ꦸ
    1.66
     fuis
    1.65
     eſſ
    1.64
     desarrollando
    1.64
     剧照
    1.64
    Activations Density 0.435%

    No Known Activations