© 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. 49739
    Prev
    Next
    INDEX
    Explanations

    refer to or reference request

    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
    ogan
    -0.11
    ç»·
    -0.10
    _MACRO
    -0.10
    rong
    -0.10
    aji
    -0.10
    å´½
    -0.09
     sẻ
    -0.09
    BOSE
    -0.09
    orca
    -0.09
    ä»Ķ
    -0.09
    POSITIVE LOGITS
    ERENCE
    0.11
    rence
    0.10
     лиÑĨ
    0.10
    (reference
    0.10
    erral
    0.09
    /reference
    0.09
    resher
    0.09
    akan
    0.09
    uge
    0.09
    reference
    0.09
    Activations Density 0.155%

    No Known Activations