© 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. 2963
    Prev
    Next
    INDEX
    Explanations

    specifying position or role

    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
    çªĺ
    -0.10
    åı»
    -0.10
    ORN
    -0.09
    å°´å°¬
    -0.09
    .intValue
    -0.08
    éĢĨè¡Į
    -0.08
    å°ıåŃIJ
    -0.08
    /oct
    -0.08
     other
    -0.08
    ä½łè¿ĺ
    -0.08
    POSITIVE LOGITS
    第
    0.15
    笫
    0.10
    ç´ł
    0.10
    :nth
    0.09
    第åįģ
    0.09
    _serial
    0.09
     booths
    0.09
     第
    0.09
     Topic
    0.08
    éĤ£ä¸Ģ
    0.08
    Activations Density 0.293%

    No Known Activations