© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 101430
    Prev
    Next
    INDEX
    Explanations

    vampire, amp, viper

    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
    
    -3.02
    轳
    -3.02
    錚
    -2.84
    辚
    -2.81
    
    -2.72
     participó
    -2.63
    
    -2.61
    顰
    -2.56
    戩
    -2.55
    咈
    -2.45
    POSITIVE LOGITS
     –
    3.44
    b
    2.91
    y
    2.77
    Our
    2.58
    t
    2.56
    This
    2.55
    Here
    2.47
    One
    2.41
    z
    2.38
    In
    2.36
    Activations Density 0.002%

    No Known Activations