© 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. 57926
    Prev
    Next
    INDEX
    Explanations

    preposition followed by 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
    Not in Any Lists

    No Comments

    Negative Logits
     meanwhile
    -0.11
    fare
    -0.10
    á»ı
    -0.10
    uft
    -0.10
    edis
    -0.09
    æĸĹ
    -0.09
    éĢļè¡Įè¯ģ
    -0.09
    缸æ¯Ķä¹ĭä¸ĭ
    -0.09
    lemn
    -0.08
     unh
    -0.08
    POSITIVE LOGITS
    åħ¶ä»ĸçļĦ
    0.11
    ella
    0.09
    æĸĻ
    0.09
    åħ¶ä»ĸ
    0.08
    åįĩåѦ
    0.08
    SDK
    0.08
    other
    0.08
    _FRIEND
    0.08
    persistent
    0.08
     ver
    0.08
    Activations Density 0.220%

    No Known Activations