© 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. 55748
    Prev
    Next
    INDEX
    Explanations

    breath and air

    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
    羣æŃ£çļĦ
    -0.09
     scar
    -0.08
    orge
    -0.08
    ç»ĥ
    -0.08
    mun
    -0.08
    çŀ»
    -0.08
    æ»ĭ
    -0.08
    ossa
    -0.08
     extrav
    -0.08
     Scar
    -0.08
    POSITIVE LOGITS
    åIJ¹
    0.19
     breath
    0.13
    对çĿĢ
    0.11
     breathed
    0.11
    åĺĺ
    0.11
    äºĨä¸Ģåı£æ°Ķ
    0.11
    æ°Ķæģ¯
    0.11
     exh
    0.11
    使åĬ²
    0.11
     blow
    0.10
    Activations Density 0.064%

    No Known Activations