© 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. 34-GEMMASCOPE-RES-131K
    7. 58526
    Prev
    Next
    INDEX
    Explanations

    error messages or code comments

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

    The neuron strongly responds to programming‐language code elements (API calls, identifiers, keywords, 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
    Vendors
    -0.68
    iações
    -0.66
     observ
    -0.66
     rabbi
    -0.65
     overlooks
    -0.64
    ylvanian
    -0.64
     Morav
    -0.62
    司令
    -0.60
    CollectionUtils
    -0.60
    сія
    -0.60
    POSITIVE LOGITS
     geheel
    0.74
    有機
    0.74
    dillo
    0.68
     Excessive
    0.67
     stads
    0.65
     شکر
    0.65
    蜂蜜
    0.64
    陲
    0.64
     energies
    0.63
     vårt
    0.63
    Activations Density 0.049%

    No Known Activations