© 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. 20096
    Prev
    Next
    INDEX
    Explanations

    listening and answering

    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¸įçŁ¥
    -0.10
    å²Ń
    -0.08
    تØŃ
    -0.08
    proc
    -0.08
    精彩
    -0.08
    .bt
    -0.08
    è·³
    -0.08
    å¯Ħ
    -0.08
    apo
    -0.08
    ä¸Ģæŀļ
    -0.08
    POSITIVE LOGITS
    åĢ¾åIJ¬
    0.20
    åIJ¬åıĸ
    0.19
    èģĨ
    0.18
    èģĨåIJ¬
    0.17
     listen
    0.17
     listened
    0.16
    .listen
    0.15
     listens
    0.15
     Listen
    0.14
    Listening
    0.14
    Activations Density 0.048%

    No Known Activations