© 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. 57768
    Prev
    Next
    INDEX
    Explanations

    Japanese place names

    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
    htdocs
    -0.09
    æĿĥ
    -0.08
    .nextElement
    -0.08
     Joined
    -0.08
    Hdr
    -0.08
    _quad
    -0.08
    icter
    -0.08
    ork
    -0.08
    çĩĶ
    -0.08
    arent
    -0.08
    POSITIVE LOGITS
     sustain
    0.09
     Witness
    0.09
    å¼¹
    0.08
    åķĨåĬ¡
    0.08
    æĤ²
    0.08
    å°±ä¸įèĥ½
    0.08
    озв
    0.08
    ä¸ĢéŨ
    0.08
    lac
    0.08
    ê·Ģ
    0.08
    Activations Density 0.138%

    No Known Activations