© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103351
    Prev
    Next
    INDEX
    Explanations

    Names followed by verbs or titles

    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_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
    During
    -1.70
     Through
    -1.60
     During
    -1.59
     and
    -1.47
     Sometimes
    -1.46
     Those
    -1.45
    Often
    -1.41
     While
    -1.41
    When
    -1.39
     梦
    -1.38
    POSITIVE LOGITS
     who
    1.68
    ↵↵↵
    1.55
     intend
    1.53
     like
    1.45
     wasn
    1.44
     lucru
    1.38
     didn
    1.38
    缀
    1.37
                  
    1.37
     sinais
    1.36
    Activations Density 0.003%

    No Known Activations