© 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. 1700
    Prev
    Next
    INDEX
    Explanations

    mul or ule

    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
    ium
    -0.11
    ening
    -0.10
    æ±¹
    -0.09
    é²Ł
    -0.09
    çīµæĮĤ
    -0.09
    èįij
    -0.09
    大åݦ
    -0.08
     innocent
    -0.08
    ãģĹãģ¾ãģĨ
    -0.08
    éĺħ读åħ¨æĸĩ
    -0.08
    POSITIVE LOGITS
    åĴĮåľ°åĮº
    0.09
    ãģĤãĤĭç¨ĭ度
    0.09
    åĭĭç«ł
    0.08
    sWith
    0.08
     unpl
    0.08
    /by
    0.08
     каÑĩ
    0.08
     Ñģобой
    0.08
    èĬ¯çīĩ
    0.08
    èµħ
    0.08
    Activations Density 0.007%

    No Known Activations