© 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. 109339
    Prev
    Next
    INDEX
    Explanations

    SVMs and parameters

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

    The neuron strongly activates on technical machine‐learning terms—especially SVM/LIBSVM identifiers and hyperparameter names (like C, gamma, grid search, 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
    ierenden
    -0.91
    issors
    -0.84
    äß
    -0.82
    )|^{
    -0.80
     المدينة
    -0.77
    atea
    -0.73
    𝖞
    -0.72
    piecz
    -0.71
     saltar
    -0.70
    Ass
    -0.70
    POSITIVE LOGITS
    Support
    1.26
     Support
    1.24
     SVC
    1.23
     SV
    1.20
    SVC
    1.16
     SVM
    1.14
     support
    1.14
    support
    1.13
    Margins
    1.10
    SV
    1.06
    Activations Density 0.018%

    No Known Activations