© 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. 58058
    Prev
    Next
    INDEX
    Explanations

    listings and headings

    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
    ivity
    -0.11
    kop
    -0.10
    erness
    -0.08
     пиÑĤаниÑı
    -0.08
    ä¹İ
    -0.08
     Ñģобой
    -0.08
    SIDE
    -0.08
    ffer
    -0.08
    bara
    -0.08
    å®ķ
    -0.08
    POSITIVE LOGITS
    ss
    0.11
    redients
    0.10
    èĬĤåģĩæĹ¥
    0.10
    olds
    0.09
    ary
    0.09
    amt
    0.09
     Stones
    0.09
    naire
    0.09
    hausen
    0.09
    reen
    0.09
    Activations Density 0.264%

    No Known Activations