© 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. 53720
    Prev
    Next
    INDEX
    Explanations

    this blog

    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
     whip
    -0.10
    ary
    -0.09
    hm
    -0.09
    igel
    -0.08
    olin
    -0.08
    ead
    -0.08
    ev
    -0.08
    ä»ĵ
    -0.08
    éĢļçĶ¨
    -0.07
    /functions
    -0.07
    POSITIVE LOGITS
    æľ¬ç½ij
    0.11
    æīĭæĦŁ
    0.10
    æijĪ
    0.10
    åĬŀ好
    0.10
    elves
    0.09
    annon
    0.09
    ripple
    0.09
     какие
    0.09
    ãģĵãģĿ
    0.09
    اÙĪ
    0.09
    Activations Density 0.107%

    No Known Activations