© 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. 48853
    Prev
    Next
    INDEX
    Explanations

    prepositions and conjunctions

    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
    åĮ¿
    -0.08
    yal
    -0.08
    è¡Įä¸ļä¸Ń
    -0.08
    ä¸Ńæĸ°
    -0.08
     fab
    -0.07
     fieldType
    -0.07
    esk
    -0.07
    adel
    -0.07
    é»ĺ
    -0.07
    osis
    -0.07
    POSITIVE LOGITS
    VRT
    0.09
    anken
    0.09
    egot
    0.09
    ä¸į妨
    0.09
    ç½Ķ
    0.09
     Preconditions
    0.09
    å¾Īæ£Ĵ
    0.09
    Ìģ
    0.09
    ynet
    0.09
    appable
    0.08
    Activations Density 0.156%

    No Known Activations