© 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. 18430
    Prev
    Next
    INDEX
    Explanations

    author and publisher of books

    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
    pole
    -0.09
    åıijå±ķ空éĹ´
    -0.09
    atum
    -0.08
    æĮĪ
    -0.08
    nic
    -0.08
    èµĦéĩij
    -0.08
    å®°
    -0.08
    è¿Ł
    -0.08
    æľĢæĸ°
    -0.08
    nik
    -0.07
    POSITIVE LOGITS
    æ¶Ķ
    0.10
    éģIJ
    0.09
    诸å¦Ĥ
    0.09
    æļ¢
    0.09
    主è¦ģåĮħæĭ¬
    0.09
     çeÅŁitli
    0.09
    åĩłåı¥
    0.09
    åĩłç§į
    0.09
    çķħéĶĢ
    0.09
    opsis
    0.09
    Activations Density 0.079%

    No Known Activations