© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112070
    Prev
    Next
    INDEX
    Explanations

    feather and subsequent words

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

    The neuron fires on any occurrence of the substring “feather” (in words like feathering, featherweight, featherstone, feather boas, 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_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
    \
    -2.92
    u
    -2.91
    ";
    -2.58
    釹
    -2.45
    le
    -2.44
    
    -2.38
    E
    -2.36
     ,‬
    -2.33
    ch
    -2.22
    H
    -2.20
    POSITIVE LOGITS
    Ꮞ
    3.13
    趼
    3.00
     и
    2.53
    阄
    2.53
     podremos
    2.38
    Krank
    2.38
    驁
    2.38
    thansa
    2.36
    你先
    2.33
     the
    2.31
    Activations Density 0.006%

    No Known Activations