© 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. 33969
    Prev
    Next
    INDEX
    Explanations

    bud and budding words

    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
    ãģĦãģĨ
    -0.11
    ï¸
    -0.09
    ingly
    -0.09
    elts
    -0.09
    eous
    -0.09
    æ¼ī
    -0.08
    ample
    -0.08
    ä¹ħ
    -0.08
    eres
    -0.08
    asion
    -0.08
    POSITIVE LOGITS
    ÑģÑĤеÑĢ
    0.11
     Budd
    0.10
    itial
    0.10
     bud
    0.10
    亲人
    0.09
    ging
    0.09
    bud
    0.09
    èĬ½
    0.09
    ĭè¯ķ
    0.09
    SHOT
    0.09
    Activations Density 0.007%

    No Known Activations