© 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. 63142
    Prev
    Next
    INDEX
    Explanations

    had a chance to

    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
    éĥ¦
    -0.09
    æľīå¤ļ
    -0.09
    aces
    -0.08
    oba
    -0.08
    å±IJ
    -0.08
    ä¸įç¦ģ
    -0.08
    ç§į
    -0.08
    axon
    -0.08
    çļĦæĶ¯æĮģ
    -0.08
    æºŁ
    -0.08
    POSITIVE LOGITS
    oka
    0.09
    lay
    0.09
    çĸĥ
    0.08
    å¾Ī好çļĦ
    0.08
    lug
    0.08
    å¹²æű
    0.08
     clown
    0.08
    è¿Ļ两个
    0.08
    çĹ
    0.08
     portion
    0.08
    Activations Density 0.053%

    No Known Activations