© 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. 10920
    Prev
    Next
    INDEX
    Explanations

    pour into container

    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
    uffers
    -0.10
    .uf
    -0.10
     пÑĢез
    -0.10
    rien
    -0.09
    amente
    -0.09
    åĨįä¹Łä¸į
    -0.09
    orrow
    -0.09
    毳
    -0.08
    ä½İ头
    -0.08
    ç¿Ĭ
    -0.08
    POSITIVE LOGITS
    缸åħ³åĨħ容
    0.11
    è£ģ
    0.09
    åīª
    0.09
    åħ¥å¢ĥ
    0.09
    /
    0.08
    ä¿İ
    0.08
    缼
    0.08
     trib
    0.08
    ä¸ĭåİ»
    0.08
    æĦıè§ģ建议
    0.08
    Activations Density 0.001%

    No Known Activations