© 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. 36006
    Prev
    Next
    INDEX
    Explanations

    very good + noun

    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
    åŀĭä¼ģä¸ļ
    -0.10
    å·´å·´
    -0.09
    ç©¿è¿ĩ
    -0.09
    ðŁı»
    -0.09
    åĵªåĦ¿
    -0.09
    æł¼æŀĹ
    -0.09
    åIJĦåĽ½
    -0.09
    ogg
    -0.08
    VICES
    -0.08
    дол
    -0.08
    POSITIVE LOGITS
     abbreviated
    0.08
    ç¬ĭ
    0.08
    nel
    0.08
    纳
    0.08
    standing
    0.08
    reso
    0.08
    ucch
    0.08
    ób
    0.08
     Mari
    0.08
     consort
    0.08
    Activations Density 0.014%

    No Known Activations