© 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. 17782
    Prev
    Next
    INDEX
    Explanations

    garden

    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
    itt
    -0.10
    itional
    -0.09
    ðŁı»
    -0.09
    vang
    -0.08
    箩
    -0.08
    sst
    -0.08
    ëł¹
    -0.08
    lop
    -0.08
    ·
    -0.08
    diag
    -0.08
    POSITIVE LOGITS
    liness
    0.10
    edException
    0.10
    ORMAL
    0.09
    iales
    0.09
    ivals
    0.09
    æĩ¿
    0.09
    æŀķ
    0.09
    Ñħи
    0.09
    æĸ°åĮº
    0.08
    mö
    0.08
    Activations Density 0.010%

    No Known Activations