© 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. 57841
    Prev
    Next
    INDEX
    Explanations

    review and preview

    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¸īåĪĨ
    -0.09
    主åĬ¨æĢ§
    -0.09
    ancell
    -0.09
    è²Į
    -0.09
    产åѦ
    -0.09
    è³´
    -0.08
    æ°¾
    -0.08
    fell
    -0.08
    anon
    -0.08
    çķ¿
    -0.08
    POSITIVE LOGITS
    åĽŀ顾
    0.20
    æĢ»ç»ĵ
    0.18
    å±ķæľĽ
    0.14
     review
    0.14
    梳çIJĨ
    0.13
    æķ´çIJĨ
    0.13
    å½Ĵ纳
    0.13
    åĽŀé¦ĸ
    0.13
    åıįæĢĿ
    0.13
    review
    0.13
    Activations Density 0.058%

    No Known Activations