© 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. 64809
    Prev
    Next
    INDEX
    Explanations

    dust and dust-related contexts

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

    The neuron detects references to particulate contaminants (e.g. “dust,” “dirt,” “soot,” 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_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
    as
    -2.89
    轳
    -2.70
    S
    -2.69
     devido
    -2.66
    na
    -2.63
    7
    -2.56
    IN
    -2.50
     Commonly
    -2.48
    R
    -2.47
    to
    -2.42
    POSITIVE LOGITS
    芣
    2.61
    蠖
    2.44
    鯪
    2.41
    垕
    2.36
    色の
    2.34
    虓
    2.33
    Ꮸ
    2.33
    蔼
    2.31
    鷓
    2.30
    珝
    2.25
    Activations Density 0.011%

    No Known Activations