© 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. 23745
    Prev
    Next
    INDEX
    Explanations

    obvious plus question, choice, or other word

    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
    جاÙĨ
    -0.09
    æ¿Ģèµ·
    -0.09
    æł¹æľ¬
    -0.09
    ched
    -0.09
    pig
    -0.08
    ä¸įå®ļ
    -0.08
    @d
    -0.08
    kir
    -0.08
    åŁºæľ¬æĥħåĨµ
    -0.08
    pite
    -0.08
    POSITIVE LOGITS
    ness
    0.14
    ly
    0.12
    mente
    0.12
     obvious
    0.12
    LY
    0.10
    赢家
    0.09
    å¥Ķ
    0.09
    NESS
    0.09
    est
    0.09
    èģĶ
    0.09
    Activations Density 0.020%

    No Known Activations