© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7280
    Prev
    Next
    INDEX
    Explanations

    more and more

    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
    metis
    -0.09
    edException
    -0.09
    edImage
    -0.09
    аÑĢаÑĤ
    -0.09
    æ³¢çī¹
    -0.09
    çĹĩ
    -0.08
    yat
    -0.08
    pek
    -0.08
    é¡¿
    -0.08
    edList
    -0.08
    POSITIVE LOGITS
    -and
    0.16
     à¹Ĩ
    0.12
    à¹Ĩ
    0.12
    ãĢħ
    0.12
    åĴĮå®ĮåĸĦ
    0.12
    _and
    0.11
    åIJĦæł·
    0.11
     ÙĪاÙĦع
    0.11
    And
    0.10
    ãģ°ãģĭãĤĬ
    0.10
    Activations Density 0.062%

    No Known Activations