© 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. 40868
    Prev
    Next
    INDEX
    Explanations

    tagged content categories

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

    This neuron detects section‐heading or metadata label patterns—i.e. short interface or header lines that use words like “on,” “theme,” “Archive,” “tagged,” etc., often followed by a colon or quotation mark.

    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
    撂
    -1.88
    踮
    -1.55
    摁
    -1.49
    饜
    -1.45
    狞
    -1.44
    踱
    -1.44
    呖
    -1.43
    摸了
    -1.40
    national
    -1.36
    Па
    -1.35
    POSITIVE LOGITS
     that
    1.97
     their
    1.95
     them
    1.83
     a
    1.80
     as
    1.67
     each
    1.33
     one
    1.27
     กรัม
    1.27
     those
    1.25
     which
    1.22
    Activations Density 0.018%

    No Known Activations