© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51021
    Prev
    Next
    INDEX
    Explanations

    Allen

    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
    大家ä¸Ģèµ·
    -0.10
    acles
    -0.10
    ç«¿
    -0.09
    inals
    -0.09
    æ¤Ģ
    -0.09
    éĿ¡
    -0.09
    ulatory
    -0.09
     ÑģоÑĤ
    -0.09
    anova
    -0.09
    acle
    -0.09
    POSITIVE LOGITS
    most
    0.10
    ÅŁt
    0.10
    ows
    0.09
    baugh
    0.09
    ness
    0.09
    ey
    0.08
    æ·±å¤Ħ
    0.08
    owed
    0.08
    çıij
    0.08
    oment
    0.08
    Activations Density 0.008%

    No Known Activations