© 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. 62837
    Prev
    Next
    INDEX
    Explanations

    purpose or direction

    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
    ä¸Ģç¬Ķ
    -0.09
    ä»İä¸Ń
    -0.09
    itics
    -0.08
     focal
    -0.08
    缼è¡Į
    -0.08
    amp
    -0.08
    èģļ
    -0.08
     fortunes
    -0.08
    误
    -0.08
    mers
    -0.08
    POSITIVE LOGITS
    åĪ«çļĦ
    0.10
     ÑģобÑģÑĤвенно
    0.09
    满äºĨ
    0.09
     chứ
    0.09
    éĽħ
    0.09
    éŀĺ
    0.09
    åĨĻçĿĢ
    0.09
    åİ»äºĨ
    0.09
    大äºĭ
    0.09
    orde
    0.08
    Activations Density 0.047%

    No Known Activations