© 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. 58295
    Prev
    Next
    INDEX
    Explanations

    extent to which

    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
    s
    -0.09
    istic
    -0.08
    -0.08
    intosh
    -0.08
     некоÑĤоÑĢÑĭе
    -0.07
    ites
    -0.07
    ìĶ©
    -0.07
    rette
    -0.07
    è¿Ļåĩłä¸ª
    -0.07
    IMUM
    -0.07
    POSITIVE LOGITS
    们éĥ½
    0.10
    rogen
    0.10
    ðŁĮ¸
    0.10
    大å¦Ī
    0.10
    æĸ¼æĺ¯
    0.09
    ally
    0.09
    iated
    0.09
    们çļĦ
    0.09
    erval
    0.09
    èµŀèµı
    0.09
    Activations Density 0.440%

    No Known Activations