© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19969
    Prev
    Next
    INDEX
    Explanations

    best practices for

    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
    hr
    -0.09
    pray
    -0.09
    æĸ¹ä¾¿
    -0.09
    çģµæ°Ķ
    -0.09
    osi
    -0.08
    翻身
    -0.08
    bane
    -0.08
    iore
    -0.08
     Giul
    -0.08
    ombo
    -0.08
    POSITIVE LOGITS
    imal
    0.09
    erno
    0.09
    æĥľ
    0.08
    Corp
    0.08
    ogen
    0.08
    actices
    0.08
    äºĪ
    0.08
    äºĪ以
    0.08
     presumption
    0.08
    änn
    0.08
    Activations Density 0.023%

    No Known Activations