© 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. 63430
    Prev
    Next
    INDEX
    Explanations

    prior art description or legal case naming

    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
    çĭ¬åħ·
    -0.09
    æİł
    -0.08
    å°ı鸣
    -0.08
    è·¯çģ¯
    -0.08
    æĭ¥
    -0.08
    鸣
    -0.08
    ç¼Ģ
    -0.08
     LM
    -0.08
    pires
    -0.08
    å±±æĿij
    -0.08
    POSITIVE LOGITS
     Hog
    0.10
    imming
    0.09
    hog
    0.09
    æı©
    0.09
     Kro
    0.09
    好è¿IJ
    0.09
    Intl
    0.08
    ARTH
    0.08
    aghan
    0.08
    ior
    0.08
    Activations Density 0.090%

    No Known Activations