Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 8619
    Prev
    Next
    INDEX
    Explanations

    copyright or creation notice

    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
     Nedir
    -1.07
     manualidades
    -1.02
     vagyok
    -0.98
     női
    -0.98
     simplemente
    -0.96
    AHULUAN
    -0.95
     คือ
    -0.93
    lè
    -0.93
     köny
    -0.92
     Arabian
    -0.91
    POSITIVE LOGITS
     this
    1.49
    This
    1.40
     follow
    1.38
     additional
    1.14
    The
    1.11
    For
    1.08
     Follow
    1.07
     stay
    1.04
    To
    1.03
     corrections
    1.03
    Activations Density 0.009%

    No Known Activations