© 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. 4540
    Prev
    Next
    INDEX
    Explanations

    to me / for me

    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
    é¾ĻéŨ
    -0.10
     Slater
    -0.09
    轧
    -0.09
    æĸĻ
    -0.09
     Siege
    -0.09
    éĢļ讯åijĺ
    -0.08
    æªĢ
    -0.08
    åĪĴ
    -0.08
    å¯¹ä½ł
    -0.08
    ivable
    -0.08
    POSITIVE LOGITS
    FUNC
    0.09
    chai
    0.08
    è¾ĩ
    0.08
    icer
    0.08
     Qed
    0.08
    å§ij
    0.08
    礼æĭľ
    0.08
     stron
    0.08
    ucha
    0.08
     VIR
    0.08
    Activations Density 0.108%

    No Known Activations