© 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. 63509
    Prev
    Next
    INDEX
    Explanations

    no, nothing, neither

    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
     Cool
    -0.08
    quired
    -0.08
    0
    -0.08
    oretical
    -0.08
    rote
    -0.08
    aged
    -0.08
    ä¸įçĽ¸ä¿¡
    -0.08
    5
    -0.07
    nx
    -0.07
    говоÑĢ
    -0.07
    POSITIVE LOGITS
    thing
    0.11
    ynes
    0.10
    æ¸ļ
    0.10
    ä¸İä¼Ĺ
    0.10
    å®ŀä¸ļ
    0.09
    åħ³äºİæĪij们
    0.09
    THING
    0.09
    å®ŀäºĭ
    0.09
    æī«ä¸Ģ
    0.09
    asti
    0.09
    Activations Density 0.041%

    No Known Activations