© 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. 34-GEMMASCOPE-RES-131K
    7. 102434
    Prev
    Next
    INDEX
    Explanations

    even if

    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
     and
    -2.28
     any
    -1.57
    iramente
    -1.49
     although
    -1.45
     نیز
    -1.38
     really
    -1.34
    quelize
    -1.30
    ("")]
    -1.30
     unless
    -1.30
     pokud
    -1.29
    POSITIVE LOGITS
     it
    2.63
     they
    2.33
     there
    2.05
     we
    1.75
     alcune
    1.60
    其他
    1.51
    他の
    1.45
     you
    1.44
     alcuni
    1.41
     некоторых
    1.40
    Activations Density 0.019%

    No Known Activations