© 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. 10-GEMMASCOPE-RES-131K
    7. 58794
    Prev
    Next
    INDEX
    Explanations

    detergent, Etiquette, caterpillar

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

    The neuron identifies words related to cleaning agents and social norms, particularly "detergent" and "etiquette".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    er
    -2.52
    鈈
    -2.42
     για
    -2.27
     Berlín
    -2.23
    
    -2.19
     conjunction
    -2.17
    鱚
    -2.17
    啃
    -2.16
    ۠
    -2.14
    ためには
    -2.13
    POSITIVE LOGITS
     you
    2.58
    6
    2.50
    2
    2.42
    4
    2.38
    i
    2.33
    ↵
    2.30
    1
    2.23
    they
    2.22
    2.19
     أيض
    2.17
    Activations Density 0.001%
    No Known Activations