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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 8155
    Prev
    Next
    INDEX
    Explanations

    identifying states and metrics

    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
    ことに
    -1.16
    _
    -0.96
     calculated
    -0.93
     consommer
    -0.89
    liğ
    -0.88
     frecuente
    -0.85
    ところ
    -0.85
    allery
    -0.85
     vindo
    -0.84
    刻意
    -0.84
    POSITIVE LOGITS
     by
    2.11
     using
    1.93
     accurately
    1.58
     anhand
    1.55
     in
    1.43
     through
    1.40
     pomocí
    1.38
     from
    1.30
     for
    1.30
     via
    1.26
    Activations Density 0.442%

    No Known Activations