© 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. 57522
    Prev
    Next
    INDEX
    Explanations

    most, best, least, happiest

    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
    éĢĤåIJĪèĩªå·±
    -0.09
    缼è¡Į
    -0.09
    失
    -0.09
    æijĬ
    -0.09
    æµģè¡ĮçļĦ
    -0.09
    å¸Ń
    -0.09
    è´«è¡Ģ
    -0.09
    代è¨Ģ
    -0.09
    è¿Ļ次
    -0.08
    å½ĵä¹ĭ
    -0.08
    POSITIVE LOGITS
    åĬŁå¤«
    0.09
    æ°¸
    0.09
    ä¹Ŀå·ŀ
    0.09
    acad
    0.09
    _contents
    0.08
    æįį
    0.08
    à¸Ńà¸Ķ
    0.08
     previews
    0.08
     amount
    0.08
    udas
    0.08
    Activations Density 0.036%

    No Known Activations