© 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. 71538
    Prev
    Next
    INDEX
    Explanations

    confirming, stabilizing, clarifying, preceding

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

    ChatGPT benchmark.com, classic, dangerous

    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
     cuyos
    0.45
     encompasses
    0.44
     overcomes
    0.44
     exceeds
    0.44
     Y
    0.43
     Du
    0.43
     Includes
    0.43
     includes
    0.42
     Well
    0.41
    Du
    0.41
    POSITIVE LOGITS
    strahlung
    0.49
     अवैध
    0.47
    烏
    0.47
    postion
    0.46
    textepsilon
    0.46
     referendum
    0.46
     Ashraf
    0.46
    streng
    0.46
    formation
    0.46
     அமைப்பு
    0.45
    Activations Density 0.001%

    No Known Activations