© 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. 31-GEMMASCOPE-2-RES-262K
    7. 136461
    Prev
    Next
    INDEX
    Explanations

    reducing bias

    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/layer_31_width_262k_l0_medium
    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
    坵
    0.50
     महासभा
    0.48
     स्पर्श
    0.47
     तापमान
    0.45
    ᄆ
    0.42
     cuáles
    0.42
     नाराजगी
    0.42
    <unused173>
    0.42
     machte
    0.42
    Smoothing
    0.41
    POSITIVE LOGITS
     in
    0.54
    W
    0.49
    ও
    0.48
     speeds
    0.47
     extract
    0.46
    l
    0.46
    N
    0.44
     scenic
    0.44
    도
    0.43
    ou
    0.43
    Activations Density 0.004%

    No Known Activations