© 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. 23485
    Prev
    Next
    INDEX
    Explanations

    glass objects and containers

    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
    åİ»çľĭ
    -0.08
    .getAction
    -0.08
    æŀļ
    -0.08
    åĮħ裹
    -0.08
    å¹ħ度
    -0.08
    ÑĤай
    -0.08
     substantive
    -0.08
    æĮ¯åħ´
    -0.08
    åĴĮåıijå±ķ
    -0.08
    é¾
    -0.08
    POSITIVE LOGITS
    ful
    0.10
     Wake
    0.10
    离
    0.09
    wipe
    0.09
    ize
    0.09
    wake
    0.09
    fully
    0.09
    澳大
    0.09
    ivec
    0.09
    pare
    0.09
    Activations Density 0.006%

    No Known Activations