© 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. 1435
    Prev
    Next
    INDEX
    Explanations

    the register

    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
    ยาà¸ĩ
    -0.09
    rear
    -0.09
    .setAction
    -0.09
    åĬ£
    -0.09
    éľİ
    -0.09
    inar
    -0.08
    ä¸Ģåħ±
    -0.08
    ä¸įæĸŃå¢ŀåĬł
    -0.08
    èįĨ
    -0.08
    åŁł
    -0.08
    POSITIVE LOGITS
    ception
    0.09
    æķij
    0.08
     triple
    0.08
    éģı
    0.08
     Mori
    0.08
    è´»
    0.07
    红线
    0.07
    ä»»èģĮ
    0.07
    èĢĺ
    0.07
     Murphy
    0.07
    Activations Density 0.047%

    No Known Activations