© 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. 38970
    Prev
    Next
    INDEX
    Explanations

    questions and perceptions

    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
    åĪ°åºķæĺ¯
    -0.09
    ç²¾ç»Ĩ
    -0.09
    èĩ³ä¸Ĭ
    -0.09
    åĬ©æĶ»
    -0.09
    çĪ¸
    -0.09
    åħ¨æĺ¯
    -0.09
    度è¿ĩ
    -0.08
    åħ¨æ°ij
    -0.08
    ç²¾éĢī
    -0.08
    åħ·æľīèī¯å¥½
    -0.08
    POSITIVE LOGITS
    عÙĤ
    0.09
    ockets
    0.09
    æĦŁè§īåĪ°
    0.09
    eras
    0.08
    ç¾ļ
    0.08
    Êİ
    0.08
    æľīæľŁ
    0.08
    raf
    0.08
    éĽ²
    0.08
    ä»ĭ
    0.08
    Activations Density 0.044%

    No Known Activations