© 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. 7343
    Prev
    Next
    INDEX
    Explanations

    health, life, quality, performance

    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
    ä¸ĸçķĮ级
    -0.09
    ovel
    -0.09
    rum
    -0.09
    åı¬åĽŀ
    -0.08
    åŁİéķĩ
    -0.08
    麾
    -0.08
    大åĨĽ
    -0.08
     Müller
    -0.08
    oki
    -0.08
    ç½Ł
    -0.08
    POSITIVE LOGITS
    -era
    0.10
     viz
    0.10
    åĬ©æīĭ
    0.10
    ä¸ĵçĶ¨
    0.10
     Kron
    0.09
     Jeh
    0.09
    å½±
    0.09
    VEC
    0.09
    -friendly
    0.09
    åĬ©çIJĨ
    0.09
    Activations Density 4.106%

    No Known Activations