© 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. 40-GEMMASCOPE-2-RES-262K
    7. 169535
    Prev
    Next
    INDEX
    Explanations

    odd numbers

    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_40_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
     دوس
    0.45
    uang
    0.43
    fälle
    0.42
    <unused24>
    0.42
    Fourth
    0.41
    hak
    0.40
     quartic
    0.40
    
    0.40
    ismod
    0.39
     lucrative
    0.39
    POSITIVE LOGITS
     odd
    0.46
     ৫১
    0.43
    奇
    0.43
     विषम
    0.40
     Iris
    0.40
     Odd
    0.39
     singers
    0.39
    浪费
    0.39
    OPTV
    0.38
    ২১
    0.38
    Activations Density 0.033%

    No Known Activations