© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 63912
    Prev
    Next
    INDEX
    Explanations

    be + adjective or noun

    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
    Negative Logits
    æĨ©
    -0.09
    åħ¬åĬ¡
    -0.09
    ican
    -0.08
    æĭĤ
    -0.08
    ä¸Ģ头
    -0.08
    ente
    -0.08
     Duty
    -0.08
    çĿ½
    -0.08
    è±ķ
    -0.08
    ä¸Ģæīĭ
    -0.08
    POSITIVE LOGITS
    åĽ´
    0.11
    mPid
    0.10
     мн
    0.10
    ADER
    0.09
    çĭĻåĩ»
    0.09
    ода
    0.09
    ç´¯
    0.09
    åľį
    0.08
    sworth
    0.08
    æ´»çĿĢ
    0.08
    Activations Density 0.030%

    No Known Activations