© 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. 3709
    Prev
    Next
    INDEX
    Explanations

    possessive noun ('s) followed by noun

    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
    ä¼ģä¸ļåıijå±ķ
    -0.11
    头çĹĽ
    -0.10
    æĽĻ
    -0.10
    è¿Ļä¹Īåģļ
    -0.10
    ä¸Ģå¹ķ
    -0.10
    åģļçļĦäºĭæĥħ
    -0.09
    大楼
    -0.09
     ÑģобÑģÑĤвенно
    -0.09
    æ¯Ķäºļ
    -0.09
    icorn
    -0.09
    POSITIVE LOGITS
    sted
    0.09
     entirety
    0.08
    èĩªå¸¦
    0.08
     surface
    0.08
    åħ¨éĥ½
    0.08
    /extensions
    0.08
    表éĿ¢
    0.08
    ine
    0.07
     metadata
    0.07
    lef
    0.07
    Activations Density 0.116%

    No Known Activations