© 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. 55370
    Prev
    Next
    INDEX
    Explanations

    options, numbers, and quantities

    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
    iban
    -0.08
    åĨįçİ°
    -0.08
     SOUR
    -0.08
    aan
    -0.08
    stral
    -0.08
     ter
    -0.08
    ALLE
    -0.08
    å§
    -0.08
    å¼ķ导
    -0.08
     EFFECT
    -0.07
    POSITIVE LOGITS
    æĪĸå¤ļ
    0.13
    èĩ³å°ij
    0.11
    oru
    0.11
    æŁIJä¸Ģ
    0.11
    ä¸įå°ijäºİ
    0.10
    Wildcard
    0.10
    æľīä¸Ģå®ļ
    0.10
    çŀ°
    0.10
    æŁIJç§į
    0.10
    å¿ħãģļ
    0.10
    Activations Density 0.111%

    No Known Activations