© 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. 58700
    Prev
    Next
    INDEX
    Explanations

    foes are, claimed:

    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
    潮湿
    -0.11
    丰å¯ĮçļĦ
    -0.10
    unken
    -0.10
     leftist
    -0.10
    éĥģ
    -0.10
    庶
    -0.10
    è±IJå¯Į
    -0.10
    ijd
    -0.09
    åıªæĺ¯ä¸Ģ个
    -0.09
    ime
    -0.09
    POSITIVE LOGITS
    çĶ¨äºº
    0.09
    çŃĭ
    0.09
    åİĭåĪ¶
    0.09
    CELER
    0.08
    çĶ¨å·¥
    0.08
    èĬŃ
    0.07
     hom
    0.07
    æű
    0.07
    设
    0.07
    çīĩéĿ¢
    0.07
    Activations Density 0.249%

    No Known Activations