© 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. 5326
    Prev
    Next
    INDEX
    Explanations

    steering, position, torque

    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
    åĢ¨
    -0.10
    éĶ¡
    -0.09
    æ£ł
    -0.09
    è°ħ
    -0.08
    çħ¤
    -0.08
     Gros
    -0.08
     acet
    -0.08
    åĢĴéľī
    -0.08
     Albert
    -0.08
    ä¸į幸
    -0.08
    POSITIVE LOGITS
    ä¼
    0.12
    åħĪéĶĭ
    0.09
    稿
    0.09
    车éģĵ
    0.09
    hel
    0.09
    urm
    0.09
    éĩijéĻµ
    0.08
    æĺ¥å¤©
    0.08
    ãģİ
    0.08
    к
    0.08
    Activations Density 0.043%

    No Known Activations