© 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. 49208
    Prev
    Next
    INDEX
    Explanations

    consultant or firm

    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
    çĽħ
    -0.09
    å¸ĺ
    -0.09
    -eyed
    -0.09
    keit
    -0.09
    едÑĮ
    -0.09
    buzz
    -0.09
    çĪ°
    -0.09
    缼åħ¸
    -0.09
    éĢĨ
    -0.08
    人éĢł
    -0.08
    POSITIVE LOGITS
    ors
    0.10
    (Constructor
    0.09
    ëĭĺ
    0.09
    åľ¨æѤä¹ĭåīį
    0.09
    firm
    0.09
    or
    0.08
    ноÑģ
    0.08
    ìĬ¬
    0.08
    èĤ¡æľ¬
    0.08
    оÑĢ
    0.08
    Activations Density 0.010%

    No Known Activations