© 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-IT
    5. 41-GEMMASCOPE-2-RES-262K
    7. 34040
    Prev
    Next
    INDEX
    Explanations

    stability and conditioning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post_all/layer_41_width_262k_l0_big
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Straßen
    0.46
    npcs
    0.45
     liệt
    0.43
     &$
    0.42
     partisan
    0.39
     muñ
    0.38
    ennes
    0.38
    殓
    0.38
     martial
    0.38
     upgrade
    0.37
    POSITIVE LOGITS
     conditioning
    0.57
     Conditioning
    0.57
    Stability
    0.56
     stability
    0.55
     стаби
    0.54
     stabilized
    0.53
    稳定
    0.52
    stability
    0.52
     Stability
    0.51
    conditioning
    0.51
    Activations Density 0.002%

    No Known Activations