© 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. 34115
    Prev
    Next
    INDEX
    Explanations

    profile sections and descriptive labels

    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
    æī¶
    -0.09
    èħ¾
    -0.09
    æ´¥
    -0.09
    æ®ĥ
    -0.09
    è¡¥
    -0.09
    è¿ĺæĺ¯è¦ģ
    -0.09
    ç§ijå°Ķ
    -0.09
    :error
    -0.08
    ç¢İ
    -0.08
    صÙĨد
    -0.08
    POSITIVE LOGITS
    ech
    0.09
    èĬ¹
    0.08
     Bucks
    0.08
    ØŃدث
    0.08
    æıī
    0.08
    èĻIJ
    0.08
    LLU
    0.07
    strup
    0.07
    avier
    0.07
     yat
    0.07
    Activations Density 0.364%

    No Known Activations