© 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. 56105
    Prev
    Next
    INDEX
    Explanations

    violinist or cello player

    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
    iron
    -0.08
    è¦ģ注æĦı
    -0.08
    ëģĶ
    -0.08
    itan
    -0.08
    饮æĸĻ
    -0.08
    utral
    -0.08
    semb
    -0.08
    SIGN
    -0.08
    大éŨ
    -0.07
    RIC
    -0.07
    POSITIVE LOGITS
    çķĮçļĦ
    0.12
    eros
    0.11
    yer
    0.10
    è¡Įéķ¿
    0.10
    trap
    0.10
    è´¨åľ°
    0.10
    æĶ¸
    0.10
    ISTS
    0.09
    atrice
    0.09
    ists
    0.09
    Activations Density 0.011%

    No Known Activations