© 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. 22381
    Prev
    Next
    INDEX
    Explanations

    subscribe to

    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
    åĩłåĪĨ
    -0.10
    istically
    -0.09
    æĭIJ
    -0.09
    mente
    -0.09
    fully
    -0.09
    ationship
    -0.08
    oted
    -0.08
    ánh
    -0.08
    zap
    -0.08
    /original
    -0.08
    POSITIVE LOGITS
    berman
    0.10
    OrUpdate
    0.10
    æ²Ĩ
    0.09
    æĻ¨æĬ¥
    0.09
    peri
    0.09
     Reviewed
    0.09
    ÐŀÐĿ
    0.09
    åĬ¼
    0.09
    åıĪæĺ¯
    0.08
    chn
    0.08
    Activations Density 0.010%

    No Known Activations