© 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. 33660
    Prev
    Next
    INDEX
    Explanations

    filler sounds and hesitations

    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
    ittings
    -0.11
    èĨ
    -0.10
    agara
    -0.09
    æıIJ款
    -0.09
     Mig
    -0.08
    esco
    -0.08
     Imam
    -0.08
    ussian
    -0.08
    [loc
    -0.08
    å°¸
    -0.08
    POSITIVE LOGITS
    æĢ»è§īå¾Ĺ
    0.10
    褪
    0.10
    å·¾
    0.10
     digest
    0.09
     scratch
    0.09
    åŁł
    0.09
    好象
    0.09
    ä¹Łåı¯èĥ½
    0.09
    åľ¨è¿ĩ渡
    0.08
     Scratch
    0.08
    Activations Density 0.022%

    No Known Activations