© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 35598
    Prev
    Next
    INDEX
    Explanations

    Diana and subsequent names

    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
    est
    -0.10
    oire
    -0.09
    è¯Ł
    -0.09
    èĥ±
    -0.09
    el
    -0.09
    家ä¼Ļ
    -0.09
    丫
    -0.09
    abled
    -0.08
    ·
    -0.08
    esh
    -0.08
    POSITIVE LOGITS
    soever
    0.10
    istic
    0.10
    á¹ĩ
    0.09
    bens
    0.09
    ../
    0.09
    AMPL
    0.08
    obil
    0.08
    .enumer
    0.08
    kest
    0.08
    รวà¸Ķ
    0.08
    Activations Density 0.006%

    No Known Activations