© 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-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 161918
    Prev
    Next
    INDEX
    Explanations

    context: Dierks Bentley, Chris Young, Jon Pardi

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

    describing specific contexts/qualities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    𝚍
    0.65
    removed
    0.54
    {(
    0.52
    𝚝
    0.52
    𝚗
    0.52
    𝚌
    0.50
    øst
    0.49
    Hlav
    0.49
    lacian
    0.49
    Remove
    0.49
    POSITIVE LOGITS
     Sancho
    0.54
     ingu
    0.46
     abolition
    0.46
     anodic
    0.46
     為
    0.46
     fave
    0.46
     Antoine
    0.45
     consape
    0.45
     szere
    0.44
     鳥
    0.44
    Activations Density 0.001%

    No Known Activations