© 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. 21357
    Prev
    Next
    INDEX
    Explanations

    features, size, presence, expression

    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¸įèī¯
    -0.10
     Krak
    -0.10
    SSERT
    -0.09
     Sachs
    -0.09
    ä¸Ńåħ±
    -0.09
    oplay
    -0.09
    æ±ł
    -0.09
    å®ł
    -0.09
    hiro
    -0.08
    è¾ľ
    -0.08
    POSITIVE LOGITS
    çī¹å¾ģ
    0.13
     features
    0.10
    åĴ¬
    0.09
    ç³»
    0.09
     feature
    0.09
    ran
    0.08
     ÑħаÑĢакÑĤеÑĢ
    0.08
    å¦Ĥæŀľæĺ¯
    0.08
    äºī
    0.08
    /features
    0.08
    Activations Density 0.259%

    No Known Activations