© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 111261
    Prev
    Next
    INDEX
    Explanations

    knowledge of

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

    The neuron fires strongly on abstract nouns denoting cognitive states of knowing or perceiving (e.g. “knowledge,” “awareness,” “perceptions,” “observations”).

    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
    3
    -1.92
    4
    -1.66
    </em>
    -1.49
     skimmed
    -1.41
    就已经
    -1.38
    对其
    -1.36
    </h3>
    -1.34
    Gé
    -1.33
     ندارد
    -1.31
     либо
    -1.30
    POSITIVE LOGITS
    𝙜
    1.78
     signifikan
    1.63
    ቍ
    1.56
    她说
    1.53
     regole
    1.49
     vitale
    1.48
     niektor
    1.48
     wartawan
    1.46
    itisation
    1.45
     fiscale
    1.45
    Activations Density 0.148%

    No Known Activations