© 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. 37571
    Prev
    Next
    INDEX
    Explanations

    state_checked

    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
    è¿ĻäºĽä¸ľè¥¿
    -0.10
    æī¶
    -0.10
    åķ¤éħĴ
    -0.10
    çħ²
    -0.09
    æĪ¿
    -0.09
    éķ¿éķ¿çļĦ
    -0.09
    ucher
    -0.09
    è£ħ
    -0.09
    éĢ¢
    -0.09
    bul
    -0.08
    POSITIVE LOGITS
    éĩį度
    0.09
    Cleanup
    0.08
    æĬ¢
    0.08
    EOS
    0.08
    alla
    0.08
    /utility
    0.08
    ç½®æį¢
    0.08
    æĢ§ä»·æ¯Ķ
    0.08
    èģĮåľº
    0.07
    æµ·åı£
    0.07
    Activations Density 0.040%

    No Known Activations