© 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. 4374
    Prev
    Next
    INDEX
    Explanations

    best of the best

    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
    çĥĽ
    -0.09
    å¢ĥ
    -0.09
    ilestone
    -0.09
    imated
    -0.09
    æĹŃ
    -0.09
     Peters
    -0.08
     Stroke
    -0.08
    éĥ¨åĪĨåľ°åĮº
    -0.08
    é«ĺå±Ĥ
    -0.08
    å±ł
    -0.08
    POSITIVE LOGITS
    ä¸ľè¥¿
    0.09
    sert
    0.09
    设置äºĨ
    0.08
    åĢĵ
    0.08
    æīĢèĥ½
    0.08
    åį§
    0.08
    æĿ±è¥¿
    0.08
    arrière
    0.08
    çļĦ带é¢Ĩ
    0.08
    _contin
    0.08
    Activations Density 0.031%

    No Known Activations