© 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. 27896
    Prev
    Next
    INDEX
    Explanations

    project

    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
    æѧ
    -0.11
    leftright
    -0.10
    Enumerator
    -0.09
     как
    -0.09
    亲
    -0.09
    以为
    -0.09
    被称为
    -0.09
     ÑıвлÑı
    -0.09
    å´ĸ
    -0.09
    ä»¥äººä¸ºæľ¬
    -0.09
    POSITIVE LOGITS
    çļĦä¸Ģéĥ¨åĪĨ
    0.12
    度è¿ĩ
    0.11
    çªģçł´åı£
    0.10
    sembl
    0.10
    çļĦä¸Ģåijĺ
    0.10
    æĬ¤èº«ç¬¦
    0.09
    æīĭ段
    0.09
    ä¸Ģéĥ¨åĪĨ
    0.09
    çľĭå¾ħ
    0.09
    ady
    0.09
    Activations Density 0.400%

    No Known Activations