© 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. 2486
    Prev
    Next
    INDEX
    Explanations

    cause incorrect behavior

    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
     stressors
    -0.90
     оказалось
    -0.85
     prelimin
    -0.85
    ulescu
    -0.84
    ierungs
    -0.84
     Baptiste
    -0.82
     Hür
    -0.81
     crudo
    -0.81
     budgetary
    -0.80
    ättern
    -0.79
    POSITIVE LOGITS
     incorrect
    1.36
     unpredictable
    1.30
     unexpected
    1.30
     behavior
    1.26
     surprising
    1.13
     behaviour
    1.13
     surprises
    1.10
    unexpected
    1.09
     unintended
    1.03
     subtle
    1.02
    Activations Density 0.192%

    No Known Activations