© 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. 84203
    Prev
    Next
    INDEX
    Explanations

    tubular and matte finishes

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

    The neuron fires on technical or descriptive adjectives that denote material, surface, or structural properties (e.g., “matte,” “tubular,” “epithelial”).

    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
    t
    -2.66
     They
    -2.52
    焄
    -2.41
    饪
    -2.38
     särskilt
    -2.28
    they
    -2.28
    
    -2.20
    ᠁
    -2.19
    ない
    -2.17
    ࿊
    -2.17
    POSITIVE LOGITS
    ﾈ
    2.58
    𖣘
    2.47
    ＼(^
    2.38
    ｺ
    2.31
    л
    2.31
    ным
    2.28
    嵛
    2.27
    妧
    2.25
    蠖
    2.22
    ：【
    2.17
    Activations Density 0.005%

    No Known Activations