© 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. 57490
    Prev
    Next
    INDEX
    Explanations

    idiomatic phrases and sayings

    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
    æݧ
    -0.09
    éĵħ
    -0.09
    Untitled
    -0.09
    ÑĭÑĤа
    -0.09
    ë
    -0.08
     Platinum
    -0.08
    wy
    -0.08
    åĺ±
    -0.08
    æĻļå¹´
    -0.08
    rowave
    -0.08
    POSITIVE LOGITS
    æĪIJè¯Ń
    0.20
     phrase
    0.14
    è¯įè¯Ń
    0.13
     phrases
    0.13
    æ¯Ķåĸ»
    0.12
    Phrase
    0.12
    è°ļ
    0.12
    è¿Ļåı¥è¯Ŀ
    0.12
    idi
    0.11
     Phrase
    0.11
    Activations Density 0.078%

    No Known Activations