© 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. 4457
    Prev
    Next
    INDEX
    Explanations

    first-person or third-person

    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
    icast
    -0.09
    зд
    -0.09
    .Publish
    -0.08
    _endian
    -0.08
    asti
    -0.08
     endian
    -0.08
    ipment
    -0.08
    Endian
    -0.08
    ansi
    -0.08
    ÙĨÙĬ
    -0.08
    POSITIVE LOGITS
    模èĮĥ
    0.10
    é³Į
    0.09
    æĦģ
    0.09
    oke
    0.08
    åĬ¨æĦŁ
    0.08
    磴
    0.08
    ÑĤелÑĮ
    0.08
    çİ«
    0.08
    书æ³ķ家
    0.08
    å°Ķæĸ¯
    0.08
    Activations Density 0.007%

    No Known Activations