© 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. 26998
    Prev
    Next
    INDEX
    Explanations

    with your/my/their

    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
    SPELL
    -0.10
    amel
    -0.10
    iting
    -0.09
    åĽĽæĺ¯
    -0.09
    æĪijåİ»
    -0.08
    emente
    -0.08
    æīĢ以æĪij
    -0.08
    è°£
    -0.08
    ä¹Ļ
    -0.08
    имв
    -0.08
    POSITIVE LOGITS
    è¿Ļå¼ł
    0.11
    åī©ä¸ĭçļĦ
    0.10
    _EXIST
    0.09
    è¿Ļ段
    0.09
    æĽ´å¤ļä¿¡æģ¯
    0.09
    è¿Ļ份
    0.09
    éĤ£ä»½
    0.09
    çĻ¾å¹´
    0.09
    è¿Ļä¹Ī大
    0.09
     newfound
    0.09
    Activations Density 0.023%

    No Known Activations