© 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. 22-GEMMASCOPE-RES-131K
    7. 71536
    Prev
    Next
    INDEX
    Explanations

    technical configuration and access

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

    The neuron is highly sensitive to named features and proper labels—i.e. explicit product or feature names, UI/menu labels, and other capitalized service or facility names.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    циональный
    -1.11
    Refrain
    -1.07
     esimerkiksi
    -1.05
     julka
    -1.04
     даже
    -1.03
     Чтобы
    -1.02
     spørs
    -1.02
    Precautions
    -1.02
    Scared
    -0.98
    idän
    -0.97
    POSITIVE LOGITS
    これは
    1.06
    ‍
    1.03
     efforts
    1.03
    ला
    0.99
     this
    0.95
     ARISING
    0.94
     bolsos
    0.93
     obligations
    0.93
     occafion
    0.91
    として
    0.90
    Activations Density 0.033%

    No Known Activations