© 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. 72346
    Prev
    Next
    INDEX
    Explanations

    face and its common collocations

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

    The neuron is highly sensitive to occurrences of the substring “face,” i.e. words beginning with or equal to “face.”

    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
    .
    -3.02
     While
    -2.50
    –
    -2.42
    ATION
    -2.41
    
    -2.30
    *
    -2.25
    啬
    -2.25
     而
    -2.22
    考えて
    -2.17
     builds
    -2.13
    POSITIVE LOGITS
    咘
    2.86
    桄
    2.83
     帅气
    2.78
    鏷
    2.63
     bunda
    2.53
    ㍑
    2.53
    𖥦
    2.47
     новые
    2.41
    躂
    2.41
    ︓
    2.39
    Activations Density 0.023%

    No Known Activations