© 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. 3800
    Prev
    Next
    INDEX
    Explanations

    percentage or range

    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
    åĴĮåıijå±ķ
    -0.12
    ä¸İåıijå±ķ
    -0.11
    èĩªçĦ¶èĢĮ
    -0.11
    åĴĮå®ĮåĸĦ
    -0.11
    ç»ıéªĮåĴĮ
    -0.10
    èĥ½åĬĽåĴĮ
    -0.10
    è¿Ļä¸ĢåĪĩ
    -0.09
    å½ĵåĪĿ
    -0.09
    æĹ¢è¦ģ
    -0.09
    assin
    -0.09
    POSITIVE LOGITS
    åĨįåĪ°
    0.11
    为æŃ¢
    0.10
    phans
    0.10
     alike
    0.09
    /or
    0.09
    ish
    0.09
     ëį
    0.09
    Ñģком
    0.09
     depending
    0.09
    åijĪçİ°åĩº
    0.08
    Activations Density 0.267%

    No Known Activations