© 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. 1803
    Prev
    Next
    INDEX
    Explanations

    informational / informative purpose

    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
    åİī
    -0.09
    iros
    -0.09
    åıªè§ģ
    -0.09
    åħ¨åªĴä½ĵ
    -0.09
    çĸµ
    -0.09
    å®ļçĤ¹
    -0.09
    å®ĿçŁ³
    -0.08
    åĬ©çIJĨ
    -0.08
    cri
    -0.08
    è¡ĮåĬ¨è®¡åĪĴ
    -0.08
    POSITIVE LOGITS
    seite
    0.12
    èĥŃ
    0.09
    rique
    0.09
    æ¨Ĺ
    0.09
     shave
    0.09
    lee
    0.08
     полов
    0.08
    UTO
    0.08
    sthrough
    0.08
    溢
    0.08
    Activations Density 0.007%

    No Known Activations