© 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. 48747
    Prev
    Next
    INDEX
    Explanations

    different kind or approach

    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
    ä¸Ģ身
    -0.09
    TEX
    -0.08
    à¸Ļาà¸Ļ
    -0.08
    fills
    -0.08
    ifu
    -0.08
    stay
    -0.08
    andle
    -0.08
    ä¸įä¸Ĭ
    -0.08
    arna
    -0.08
     best
    -0.08
    POSITIVE LOGITS
     altogether
    0.15
    than
    0.13
     than
    0.13
    _than
    0.13
    iable
    0.12
    -than
    0.12
     kinds
    0.11
    å®Įåħ¨ä¸įåIJĮ
    0.11
    æĪĽ
    0.10
     THAN
    0.10
    Activations Density 0.081%

    No Known Activations