© 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. 39425
    Prev
    Next
    INDEX
    Explanations

    citations like Abrams et al

    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
    uur
    -0.09
    orate
    -0.09
    joy
    -0.09
     Wilkinson
    -0.08
    heet
    -0.08
    ë©ľ
    -0.08
    ç¼ĺ
    -0.08
    inals
    -0.08
    /legal
    -0.08
    ease
    -0.08
    POSITIVE LOGITS
    isse
    0.10
    æĢ§çĸ¾çĹħ
    0.09
    åľ°å¸¦
    0.09
    寨
    0.09
    带åĬ¨
    0.09
    venient
    0.09
    asan
    0.09
    ooth
    0.09
    碾
    0.09
    Ñıм
    0.09
    Activations Density 0.007%

    No Known Activations