© 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. 30796
    Prev
    Next
    INDEX
    Explanations

    enough

    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
    _VO
    -0.10
     OPTIONAL
    -0.09
     ips
    -0.09
    obra
    -0.09
    å¥Ĺè£ħ
    -0.09
     IPS
    -0.09
    ificent
    -0.09
    æ¯Ķè¾ĥå¤ļ
    -0.08
    tors
    -0.08
    ipse
    -0.08
    POSITIVE LOGITS
    ä¸Ģç¬Ķ
    0.09
    ä¸Ģ家
    0.08
    ä½łæĥ³
    0.08
    uve
    0.08
    vak
    0.08
    Invoke
    0.08
    è¿ľ
    0.08
    lob
    0.08
    oji
    0.08
    icap
    0.08
    Activations Density 0.021%

    No Known Activations