© 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. 35504
    Prev
    Next
    INDEX
    Explanations

    logic programming or planning

    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
     вÑĭпÑĥÑģ
    -0.10
    æİ·
    -0.10
    å¸
    -0.10
    anus
    -0.09
    zego
    -0.09
    åĴ½
    -0.09
    åĩ»
    -0.09
    _strike
    -0.09
    çĶ¨å¾®ä¿¡
    -0.09
    ä¸Ģ对
    -0.08
    POSITIVE LOGITS
    çĨı
    0.10
    è¡ĮåĬ¨
    0.08
    ortal
    0.08
    åİ®
    0.08
    ugin
    0.08
    éĵ®
    0.08
     planning
    0.08
     Lang
    0.08
     Planning
    0.08
    _alignment
    0.08
    Activations Density 0.107%

    No Known Activations