© 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. 63315
    Prev
    Next
    INDEX
    Explanations

    jump up and down

    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
    éĿ¢åŃIJ
    -0.09
    ILON
    -0.09
    ames
    -0.08
    enos
    -0.08
    iming
    -0.08
    oming
    -0.08
    hell
    -0.08
    UIScreen
    -0.07
    cket
    -0.07
    uib
    -0.07
    POSITIVE LOGITS
    éļĹ
    0.10
    èĻIJå¾ħ
    0.09
    åĪ°æŶåĢĻ
    0.09
    åĨ·ç¬ij
    0.09
    æĬĬè¿Ļ个
    0.09
    祧
    0.09
    æĺİå¹´
    0.09
    éĤ£ç§į
    0.09
    诸侯
    0.09
    妾
    0.09
    Activations Density 0.079%

    No Known Activations