© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 93787
    Prev
    Next
    INDEX
    Explanations

    there exists such that

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

    The neuron is picking out specialized mathematical terminology and technical jargon (e.g. names of concepts like “irrational,” “nilpotent,” “subsequence,” “Markov,” “quaternion,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     to
    -1.00
     or
    -1.00
    нять
    -0.99
    éry
    -0.98
     در
    -0.91
     a
    -0.90
     lavish
    -0.90
    齎
    -0.88
     necessário
    -0.88
     Nachteile
    -0.87
    POSITIVE LOGITS
     such
    6.63
    such
    4.25
    Such
    3.80
     таких
    3.28
     Such
    3.19
     SUCH
    3.14
     такие
    3.13
     такой
    3.09
     такое
    2.88
     такую
    2.75
    Activations Density 0.124%

    No Known Activations