© 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. 32980
    Prev
    Next
    INDEX
    Explanations

    may render, may have, need to, capable of

    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
     zwar
    -0.12
    èĻ½çĦ¶
    -0.11
     but
    -0.11
    åĽºçĦ¶
    -0.11
    ä¸įä»ħæĺ¯
    -0.11
    ä¸įä»ħä»ħæĺ¯
    -0.11
    ä¸įä»ħ
    -0.10
     но
    -0.10
    ä¸įåıªæĺ¯
    -0.10
    ä¸įä»ħè¦ģ
    -0.10
    POSITIVE LOGITS
     nonetheless
    0.25
     nevertheless
    0.23
    åį´æĺ¯
    0.17
    çħ§æł·
    0.16
    ãģĿãĤĮãģ§ãĤĤ
    0.15
    åį»
    0.14
    ä»įçĦ¶
    0.13
    ä»įæľī
    0.13
    Nevertheless
    0.13
    åį´
    0.13
    Activations Density 0.234%

    No Known Activations