© 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. 1919
    Prev
    Next
    INDEX
    Explanations

    requests for explanation

    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
    æĸĹäºī
    -0.11
    åĢ¦
    -0.10
    粤
    -0.09
    .normalized
    -0.09
    æ´»
    -0.09
    oul
    -0.09
     Bra
    -0.08
    èĻļ
    -0.08
    èĦĨ
    -0.08
    ÑĪе
    -0.08
    POSITIVE LOGITS
    è°Īè°Ī
    0.16
    ä»ĭç»įä¸Ģä¸ĭ
    0.13
    è°Ī
    0.10
    olib
    0.10
    çķ¿
    0.09
     briefly
    0.09
    仪
    0.09
    ắng
    0.09
    åĮħåĮħ
    0.09
    çĪ·
    0.09
    Activations Density 0.042%

    No Known Activations