© 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. 12050
    Prev
    Next
    INDEX
    Explanations

    legal testimony and defense

    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
    argento
    -0.92
     läh
    -0.85
    erah
    -0.84
     geleden
    -0.83
    dimer
    -0.82
    Says
    -0.81
    QD
    -0.79
     kök
    -0.79
     هوای
    -0.78
     turun
    -0.77
    POSITIVE LOGITS
     jurors
    1.90
     defense
    1.54
     testified
    1.47
     witness
    1.39
     defence
    1.34
     prosecutor
    1.34
     testifying
    1.30
     repeatedly
    1.23
     prosecution
    1.23
     later
    1.20
    Activations Density 0.018%

    No Known Activations