© 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. 35063
    Prev
    Next
    INDEX
    Explanations

    library prefix

    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
    ä¸Ģç¬Ķ
    -0.10
    /or
    -0.09
    è·ĭ
    -0.09
    avel
    -0.08
    ç¬Ļ
    -0.08
    /OR
    -0.08
    iten
    -0.08
    ä¿ļ
    -0.08
     MAS
    -0.08
    åĥµå°¸
    -0.08
    POSITIVE LOGITS
    bie
    0.11
    çĦ¶æĺ¯
    0.10
    RARY
    0.10
    raries
    0.10
    CEED
    0.09
    èĩªåľ¨
    0.09
    quit
    0.09
    ../../../
    0.09
    rarian
    0.09
    erator
    0.09
    Activations Density 0.022%

    No Known Activations