© 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. 32851
    Prev
    Next
    INDEX
    Explanations

    Horizontally or horiz-

    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
    ively
    -0.11
    erval
    -0.09
    oretical
    -0.09
    ÙĬÙĬÙĨ
    -0.09
    è°Ł
    -0.09
    ¹
    -0.09
    ãģ²ãģ¨
    -0.08
    adera
    -0.08
    arov
    -0.08
    Ń
    -0.08
    POSITIVE LOGITS
    izontally
    0.11
    izons
    0.10
    ä¸ĢåĽ¢
    0.10
    人éĢī
    0.09
    izon
    0.09
    ожно
    0.09
    rible
    0.09
    ignal
    0.09
    ribly
    0.08
    idon
    0.08
    Activations Density 0.014%

    No Known Activations