© 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. 86384
    Prev
    Next
    INDEX
    Explanations

    rush forward into

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

    The neuron flags relatively uncommon content words (i.e. non‐stopwords) such as verbs and specific nouns rather than high-frequency function words.

    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
    m
    -2.28
    providedIn
    -2.14
    </h3>
    -2.06
    We
    -1.98
    for
    -1.98
     для
    -1.98
    u
    -1.94
    ou
    -1.94
     contemporáneo
    -1.90
    ｎ
    -1.87
    POSITIVE LOGITS
     mergul
    1.73
    Ꮈ
    1.60
     îna
    1.56
     We
    1.54
     uniquement
    1.52
     pergunt
    1.52
    籥
    1.52
    惭
    1.51
    狁
    1.50
    想法
    1.48
    Activations Density 0.003%

    No Known Activations