© 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. 22-GEMMASCOPE-RES-131K
    7. 34470
    Prev
    Next
    INDEX
    Explanations

    acronyms followed by slash or hyphen

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

    The neuron activates on numeric tokens and measurement values (e.g., numbers with units or quantitative data) in scientific contexts.

    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
     sowie
    -0.89
     或
    -0.88
    以及
    -0.87
    或
    -0.83
     用
    -0.83
    Recommend
    -0.81
     oraz
    -0.81
    甚至
    -0.81
     bes
    -0.80
    健康
    -0.79
    POSITIVE LOGITS
    ținut
    1.05
    Ƕ
    0.98
    韭
    0.98
    ⮕
    0.90
    同士
    0.89
    一二
    0.89
    aminan
    0.87
    튿
    0.85
    obé
    0.85
    urations
    0.84
    Activations Density 0.084%

    No Known Activations