© 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. 4113
    Prev
    Next
    INDEX
    Explanations

    receiving mail/feedback

    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
    çļĦä¸Ģ项
    -0.09
    ä¸Ģç»Ħ
    -0.09
    ä¸ĢæĹı
    -0.09
    ament
    -0.09
    AIM
    -0.09
    .easy
    -0.08
    åĤĴ
    -0.08
    åĴĮä»ĸçļĦ
    -0.08
     ob
    -0.08
    çļĦ形象
    -0.08
    POSITIVE LOGITS
    åıįé¦Ī
    0.12
    åijĬè¯īæĪij
    0.10
    å±ħçĦ¶
    0.10
    使æĪij
    0.10
    åĽº
    0.10
    讶
    0.09
    QUIRES
    0.09
     feedback
    0.09
    æĬĬæĪij
    0.09
    feedback
    0.09
    Activations Density 0.067%

    No Known Activations