© 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. 58809
    Prev
    Next
    INDEX
    Explanations

    crist, defect, attribute

    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
    çĹĺ
    -0.09
    é»ĺ
    -0.09
    оÑģÑĮ
    -0.09
    ede
    -0.09
    egas
    -0.09
    ose
    -0.08
    é»į
    -0.08
    å·½
    -0.08
    mapping
    -0.08
    itters
    -0.08
    POSITIVE LOGITS
    ially
    0.11
    HEN
    0.09
    emonic
    0.09
    gers
    0.08
    ARN
    0.08
    åĵ¥ä»¬
    0.08
    éĥľ
    0.08
     Liberty
    0.08
    -animate
    0.08
    å·¢
    0.08
    Activations Density 0.001%

    No Known Activations