© 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. 36159
    Prev
    Next
    INDEX
    Explanations

    knock or tap on door

    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
    èººåľ¨
    -0.11
    arro
    -0.09
    对æİ¥
    -0.09
    æĺİæľĹ
    -0.09
    æľ«
    -0.09
    丰åİļ
    -0.08
    aro
    -0.08
    缴éĢļ车
    -0.08
    aleigh
    -0.08
    ynth
    -0.08
    POSITIVE LOGITS
    æķ²
    0.11
    åı©
    0.10
     knock
    0.10
     tap
    0.09
     knocking
    0.09
    -door
    0.09
    iens
    0.09
    UserCode
    0.09
    ä¸Ĭå¸Ŀ
    0.09
    æĪIJ为ä¸ŃåĽ½
    0.09
    Activations Density 0.004%

    No Known Activations