© 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. 32542
    Prev
    Next
    INDEX
    Explanations

    track of

    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
    veau
    -0.10
    æĪijçŁ¥éģĵ
    -0.10
    /ion
    -0.09
    ç«ĭåį³
    -0.09
    æĸĻ
    -0.09
    åij¤
    -0.08
    .tm
    -0.08
    rary
    -0.08
     brakes
    -0.08
    trap
    -0.08
    POSITIVE LOGITS
    æ¯ĶåĪĨ
    0.10
    ä¸įæ¸ħ
    0.09
    ãģĬå¾Ĺ
    0.09
     gist
    0.08
    æĶĴ
    0.08
    ä¸ī代
    0.08
    buch
    0.08
    失èIJ½
    0.08
    /control
    0.08
    çĪ·
    0.08
    Activations Density 0.025%

    No Known Activations