© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 27081
    Prev
    Next
    INDEX
    Explanations

    marshmallow or marshland

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ivities
    -0.09
    ivity
    -0.09
    á»ħ
    -0.09
    æĢ»å±Ģ
    -0.08
    éľľ
    -0.08
    ä¸ĭ乡
    -0.08
    åijIJ
    -0.08
    ì²Ļ
    -0.08
    审
    -0.08
    社ä¼ļèµĦæľ¬
    -0.08
    POSITIVE LOGITS
    mallow
    0.13
    è¡·
    0.11
    èĩ¾
    0.10
    engers
    0.10
    andise
    0.09
    iest
    0.09
    ensen
    0.09
    èĮ«èĮ«
    0.09
    bourne
    0.09
    agne
    0.08
    Activations Density 0.004%

    No Known Activations