© 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. 25910
    Prev
    Next
    INDEX
    Explanations

    combinatorial analysis and optimization

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

    The neuron activates strongly on words containing the root “combin” (e.g. combinatorics, combinatorial, combinatorialization).

    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
     Remake
    -0.82
     Serializable
    -0.79
     Latter
    -0.73
     Zayn
    -0.70
    собенно
    -0.70
    QByteArray
    -0.69
     DPM
    -0.69
    👿
    -0.69
     Lach
    -0.68
    мото
    -0.68
    POSITIVE LOGITS
     Curtis
    0.73
    Mora
    0.71
    xad
    0.69
    Curtis
    0.65
    icated
    0.63
     frost
    0.63
    ुर
    0.63
     validated
    0.62
    validated
    0.61
     miljö
    0.61
    Activations Density 0.063%

    No Known Activations