© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 9391
    Prev
    Next
    INDEX
    Explanations

    listing and checking

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

    The neuron fires on words written in non‐Latin scripts—especially Japanese or Chinese character sequences (and similar Unicode tokens).

    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
     was
    -1.72
     Если
    -1.56
     IBOutlet
    -1.54
     If
    -1.34
    ativement
    -1.33
     ако
    -1.29
    expiresIn
    -1.26
     если
    -1.25
    <bos>
    -1.25
     BorderLayout
    -1.23
    POSITIVE LOGITS
     motoru
    1.59
    isbury
    1.53
     doigt
    1.34
    futbolista
    1.32
     mojej
    1.32
    larım
    1.30
     моего
    1.30
     vectorielle
    1.30
     Inscrivez
    1.29
     nôtre
    1.27
    Activations Density 0.003%

    No Known Activations