© 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. 57024
    Prev
    Next
    INDEX
    Explanations

    "only" followed by punctuation

    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
    åİī
    -0.09
    讲究
    -0.09
    ç»ķ
    -0.08
    æľīåĪ©
    -0.08
    ru
    -0.08
    VERSE
    -0.08
    mgr
    -0.08
    sur
    -0.08
    rä
    -0.08
    reject
    -0.08
    POSITIVE LOGITS
    IELDS
    0.09
    ç¼Ĩ
    0.09
    åįİ举
    0.09
    çIJħ
    0.09
    èĥ¯
    0.09
    å°ıåĦ¿
    0.08
    å°ıå¾®
    0.08
    èĤ¡ä¸ľå¤§ä¼ļ
    0.08
    templ
    0.08
    èĥ¡åŃIJ
    0.08
    Activations Density 0.006%

    No Known Activations