© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20492
    Prev
    Next
    INDEX
    Explanations

    legal declarations

    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
    structions
    -0.09
    ching
    -0.09
    ä½łçŁ¥éģĵ
    -0.09
    atories
    -0.09
    _ns
    -0.09
    otron
    -0.08
    ienes
    -0.08
    ÑģÑĤав
    -0.08
    yro
    -0.08
    byn
    -0.08
    POSITIVE LOGITS
    ç®Ģ称
    0.11
    ç½ijé¦ĸ页
    0.09
    éĻ¤æѤ
    0.09
    åīįèĢħ
    0.09
    éī´äºİ
    0.08
    æĪ¿è½¦
    0.08
    åIJİèĢħ
    0.08
    è¿ĺæĺ¯è¦ģ
    0.08
    å¤ļè¾¾
    0.08
    acks
    0.07
    Activations Density 6.457%

    No Known Activations