© 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. 48030
    Prev
    Next
    INDEX
    Explanations

    attention grabbing

    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
    çĮĹ
    -0.11
    Č
    -0.10
    å°ı康社ä¼ļ
    -0.10
    à°¬
    -0.09
    ä¸Ĭæµ·è¯ģåĪ¸
    -0.09
    æĮ½åĽŀ
    -0.09
    _IOS
    -0.08
    hand
    -0.08
    outes
    -0.08
    ä¸Ĭåįĥ
    -0.08
    POSITIVE LOGITS
    çľ¼çIJĥ
    0.11
    entrant
    0.09
    eler
    0.09
    å±ı
    0.09
    çŀ¥
    0.08
    ATCH
    0.08
    shal
    0.08
     Fraser
    0.08
    ç»ıå¼Ģ
    0.08
    åįģ足
    0.08
    Activations Density 0.009%

    No Known Activations