© 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. 34-GEMMASCOPE-RES-131K
    7. 92772
    Prev
    Next
    INDEX
    Explanations

    legal, technical, medical document sections

    np_acts-logits-general · gemini-2.5-flash-lite

    URLs and file paths in technical documentation or code comments.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
     some
    -0.96
     despite
    -0.93
     specific
    -0.93
     also
    -0.90
     other
    -0.90
     after
    -0.90
     while
    -0.88
     it
    -0.87
     certainly
    -0.87
     when
    -0.85
    POSITIVE LOGITS
     Atención
    1.19
    Iowa
    1.14
     Iowa
    1.11
     Dinas
    1.03
     chaves
    0.98
     Moines
    0.96
    ípio
    0.95
     Mexique
    0.94
     numerus
    0.91
    macOS
    0.91
    Activations Density 0.000%

    No Known Activations