© 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. 62103
    Prev
    Next
    INDEX
    Explanations

    frequency of occurrences

    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
    ะ
    -0.10
    flix
    -0.10
    swer
    -0.09
    åĬŁå¤«
    -0.09
    ÑĢок
    -0.09
    istically
    -0.08
    room
    -0.08
     شأÙĨ
    -0.08
    ä¸Ģç§į
    -0.08
    çļĦæĸ¹å¼ıæĿ¥
    -0.08
    POSITIVE LOGITS
    째
    0.11
    éĶ²
    0.10
    ylvania
    0.09
    人çĶŁçļĦ
    0.09
    STDOUT
    0.09
    dre
    0.09
    åĽ½åĬ¡
    0.09
    å½»
    0.09
    è¯Ĭ
    0.09
    avo
    0.09
    Activations Density 0.029%

    No Known Activations