© 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. 22-GEMMASCOPE-RES-131K
    7. 9403
    Prev
    Next
    INDEX
    Explanations

    pros and cons

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

    The neuron primarily detects numeric tokens or ratings (e.g. digits, numbers, and scores).

    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
     mendatang
    -1.21
    之际
    -1.16
     revisor
    -1.16
     with
    -1.14
     Lycée
    -1.10
    ಆ
    -1.10
     ставить
    -1.08
    七月
    -1.06
     FATHER
    -1.06
     Bibliothèque
    -1.05
    POSITIVE LOGITS
    3
    1.07
     manchen
    1.05
    どうやって
    1.05
     amazing
    1.04
    Ekster
    1.04
    LOAT
    1.03
     vs
    1.03
     alguna
    1.02
    2
    1.02
     .,
    0.99
    Activations Density 0.001%

    No Known Activations