© 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. 23652
    Prev
    Next
    INDEX
    Explanations

    . UR

    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
    å¾Ģ
    -0.08
    (typ
    -0.08
    .Objects
    -0.08
    CESS
    -0.08
    æĬ±çĿĢ
    -0.08
    critical
    -0.07
    quel
    -0.07
    ä¸Ģéģĵ
    -0.07
    ç½ijåıĭ
    -0.07
    âĶĢ
    -0.07
    POSITIVE LOGITS
    笳
    0.11
    otron
    0.11
    Disallow
    0.10
    çĻ»ä¸Ĭ
    0.09
    è¾ľ
    0.09
    ãi
    0.09
    éĺ²
    0.09
    .UR
    0.09
    è¤Ĭ
    0.09
    atre
    0.09
    Activations Density 0.082%

    No Known Activations