© 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. 34622
    Prev
    Next
    INDEX
    Explanations

    exp followed by "ansion", "anding", "ession", "ressed", "at", "edit"

    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
    ../../../
    -0.12
    umeric
    -0.10
    pped
    -0.09
    ì¦Ī
    -0.09
    аÑĢа
    -0.09
    ãĥ£
    -0.09
    ãģĹãģ¾ãģĨ
    -0.09
    à¹ģà¸Ī
    -0.09
    iffs
    -0.08
    á»ĩt
    -0.08
    POSITIVE LOGITS
    ansion
    0.14
    ively
    0.12
    erts
    0.11
    lore
    0.11
    /exp
    0.11
    loit
    0.11
    RESSION
    0.10
     thụ
    0.10
    anding
    0.10
    ONENT
    0.10
    Activations Density 0.052%

    No Known Activations