© 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. 18105
    Prev
    Next
    INDEX
    Explanations

    city and state 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
    lier
    -0.09
     Sent
    -0.09
    缱
    -0.08
     Dank
    -0.08
    ÙħصÙĦ
    -0.08
    æĬķ注
    -0.08
    éĢģåİ»
    -0.08
     Nacht
    -0.08
    èĮ±
    -0.08
    åĸľ
    -0.08
    POSITIVE LOGITS
    说æĪij
    0.10
    æŃ£æĺ¯åĽłä¸º
    0.09
    hänge
    0.09
    ä¸Ģå®ļç¨ĭ度
    0.09
    rez
    0.09
    æį©
    0.08
    .bp
    0.08
    å½·
    0.08
    etz
    0.08
    ValueType
    0.08
    Activations Density 0.191%

    No Known Activations