© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22241
    Prev
    Next
    INDEX
    Explanations

    given that

    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
    å¦Ļ
    -0.09
    _void
    -0.09
     знаÑĩиÑĤ
    -0.09
    ched
    -0.09
    pan
    -0.09
    ól
    -0.08
    ãģ»ãģĨ
    -0.08
    åĵĩ
    -0.08
    CHAN
    -0.08
    ä¸įå®ļ
    -0.08
    POSITIVE LOGITS
    è¿Ļç§įæĥħåĨµ
    0.11
    ç§įç§į
    0.11
    å®ŀéĻħæĥħåĨµ
    0.10
    åħ¶æīĢ
    0.10
    è¿ĺæĺ¯å¾Ī
    0.09
    æѤåĪ»
    0.09
    eci
    0.08
    å®ĥæĺ¯
    0.08
    ectl
    0.08
     circumstances
    0.08
    Activations Density 0.029%

    No Known Activations