© 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. 31197
    Prev
    Next
    INDEX
    Explanations

    I want / need / you want / we want

    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
     enough
    -0.08
     only
    -0.08
    ä¸įä¸Ģå®ļ
    -0.07
    FER
    -0.07
    ä¸Ģå®ļæĺ¯
    -0.07
    est
    -0.07
    ä¹ŁåĽłæѤ
    -0.07
    æĬ¬å¤´
    -0.07
    çĭ¬
    -0.07
     Emer
    -0.07
    POSITIVE LOGITS
    åĽŀäºĭ
    0.13
    /her
    0.12
    åģļçļĦäºĭ
    0.11
    /from
    0.11
    åģļçļĦäºĭæĥħ
    0.11
    ç½ij讯
    0.10
    agini
    0.10
    ourke
    0.10
    /she
    0.10
    ses
    0.10
    Activations Density 0.181%

    No Known Activations