© 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. 38506
    Prev
    Next
    INDEX
    Explanations

    mad, anger, Midas, Hatter

    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
     disaster
    -0.73
    ା
    -0.73
    抱着
    -0.71
     focus
    -0.69
     feedback
    -0.68
     starting
    -0.68
     Gegend
    -0.67
     chir
    -0.66
     sentidos
    -0.66
     contingency
    -0.66
    POSITIVE LOGITS
     MAD
    0.90
    MAD
    0.84
    ピース
    0.83
    Mad
    0.82
     Anger
    0.81
     Hatter
    0.80
     Mad
    0.80
     recensione
    0.80
    ANGER
    0.77
     Midas
    0.77
    Activations Density 0.032%

    No Known Activations