© 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. 1623
    Prev
    Next
    INDEX
    Explanations

    hers, herschel, hershey, hersberger, wallin, rsync

    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
    istik
    -0.10
    ä»įæľī
    -0.09
    å¸ĪäºĭåĬ¡æīĢ
    -0.09
    илÑģÑı
    -0.09
    OMIC
    -0.09
    ç¨ŀ
    -0.09
    ìļĶ
    -0.09
    GINE
    -0.09
    ики
    -0.09
    ruz
    -0.09
    POSITIVE LOGITS
    ness
    0.11
    enal
    0.11
    ingly
    0.09
    unken
    0.09
    aldo
    0.09
    ival
    0.08
    ç¿ķ
    0.08
    夫
    0.08
    cheid
    0.08
    ellt
    0.08
    Activations Density 0.040%

    No Known Activations