© 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. 22770
    Prev
    Next
    INDEX
    Explanations

    SOOOO, soo, y, SO, o

    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
    bies
    -0.10
    åĦ¿
    -0.09
    æıŃéľ²
    -0.08
    åĺ´è§Ĵ
    -0.08
    andal
    -0.08
    piler
    -0.08
     Smash
    -0.08
    æĽ´ä½ķåĨµ
    -0.08
    ´
    -0.08
    bi
    -0.08
    POSITIVE LOGITS
    eva
    0.11
    å¾Īå¤ļæľĭåıĭ
    0.09
    headed
    0.09
    ommen
    0.09
    èĽ±
    0.09
    æīĵæŃ»
    0.09
    åľ¨æĦı
    0.08
    anness
    0.08
    erox
    0.08
    ç¼®
    0.08
    Activations Density 0.003%

    No Known Activations