© 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. 19352
    Prev
    Next
    INDEX
    Explanations

    parentheses and periods

    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
    æį¢ä¸ª
    -0.09
    standing
    -0.09
    åıijå±ķæľºéģĩ
    -0.09
    åĬŀ好
    -0.08
    æ±
    -0.08
    åĭ¿
    -0.08
    iss
    -0.08
    ISS
    -0.08
    åįıåĬŀ
    -0.08
    åıijå±ķçIJĨ念
    -0.08
    POSITIVE LOGITS
    RARY
    0.09
    è¿Ļå°±
    0.09
    .pretty
    0.09
    æĿ¥åĽŀ
    0.08
    inness
    0.08
    auga
    0.08
    inces
    0.08
    æĴŃ
    0.08
    _rl
    0.08
    èĶ
    0.08
    Activations Density 0.192%

    No Known Activations