© 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. 55067
    Prev
    Next
    INDEX
    Explanations

    wear and tear or strictly prohibited

    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
    zos
    -0.09
    sy
    -0.09
    åįĬå¤ľ
    -0.09
    éķ¿è¿ľ
    -0.08
    è¾Ľ
    -0.08
    adora
    -0.08
    ç»ıåİĨè¿ĩ
    -0.08
    æijĬ
    -0.08
    éº
    -0.08
    å®ŀæķĪ
    -0.08
    POSITIVE LOGITS
    erk
    0.11
    porto
    0.10
    orpor
    0.09
    á»Ļc
    0.09
    tern
    0.09
    _FUNC
    0.09
     poczÄħtk
    0.08
    åľ°äº§
    0.08
    stry
    0.08
    æŃ¤ä¸¾
    0.08
    Activations Density 0.007%

    No Known Activations