© 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. 11484
    Prev
    Next
    INDEX
    Explanations

    themes and topics

    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
     IsNot
    -0.09
    arding
    -0.09
    ä¸ĢåłĨ
    -0.08
    icy
    -0.08
    igos
    -0.08
    )null
    -0.08
    igo
    -0.08
    è¿«ä¸įåıĬå¾ħ
    -0.08
     Scr
    -0.08
    AndView
    -0.08
    POSITIVE LOGITS
    贯穿
    0.16
    åħ¨å¤©
    0.13
     thread
    0.12
    -thread
    0.12
     throughout
    0.12
    主线
    0.11
     threads
    0.11
    åŃĺåľ¨äºİ
    0.11
    thread
    0.11
    _THREAD
    0.11
    Activations Density 0.032%

    No Known Activations