© 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. 39301
    Prev
    Next
    INDEX
    Explanations

    backdrop or background

    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
    liness
    -0.10
    ä¸Ģ度
    -0.09
    ä¸ĢæŃ¥
    -0.09
    æ¸ħçĪ½
    -0.09
    ehr
    -0.09
    ä¸Ģè¶Ł
    -0.09
    _NOP
    -0.09
    coin
    -0.09
    ish
    -0.09
    owns
    -0.09
    POSITIVE LOGITS
     backdrop
    0.09
     nhiêu
    0.09
    秦çļĩ
    0.09
    ittel
    0.09
    身å¿ĥ
    0.09
    æijĺç¼ĸ
    0.09
    abay
    0.08
    严峻
    0.08
    .exit
    0.08
     Setter
    0.08
    Activations Density 0.007%

    No Known Activations