© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 117478
    Prev
    Next
    INDEX
    Explanations

    phrases following punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     perché
    -1.43
    そのため
    -1.35
    <bos>
    -1.35
    なので
    -1.27
     tandis
    -1.23
    
    -1.17
     weil
    -1.16
    があり
    -1.15
    Доброго
    -1.13
     потому
    -1.12
    POSITIVE LOGITS
     in
    2.48
     or
    2.30
     from
    2.00
     without
    1.99
     through
    1.98
     via
    1.92
     at
    1.88
     on
    1.88
     and
    1.84
     for
    1.80
    Activations Density 0.074%

    No Known Activations