© 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. 63379
    Prev
    Next
    INDEX
    Explanations

    any other

    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
     беÑģплаÑĤно
    -0.09
    æĢİä¹Īåı¯èĥ½
    -0.09
    å¨ĥå¨ĥ
    -0.08
    æŁIJç§į
    -0.08
    æĺ¯ä»Ģä¹Ī
    -0.08
    steam
    -0.08
    frag
    -0.08
    ermal
    -0.08
    ä¸Ģå®ļä¼ļ
    -0.08
    ç»Ĭ
    -0.08
    POSITIVE LOGITS
    åīįèĢħ
    0.11
    çıł
    0.10
    åıĸãĤĬ
    0.09
    ä¸ĢåĪ»
    0.09
    深交
    0.09
    _TC
    0.09
    åį´åıĪ
    0.09
    _mD
    0.09
    oster
    0.09
     Blackburn
    0.08
    Activations Density 0.007%

    No Known Activations