© 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. 49113
    Prev
    Next
    INDEX
    Explanations

    naked + descriptive noun

    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
    keeper
    -0.10
    keepers
    -0.10
    ÃŁen
    -0.10
    åĶ§
    -0.09
    åı¯æĥľ
    -0.09
    ฯ
    -0.09
    ingly
    -0.08
    esh
    -0.08
    aub
    -0.08
    æĤ¸
    -0.08
    POSITIVE LOGITS
    ness
    0.16
    NESS
    0.11
    ../../../
    0.11
    ly
    0.10
    çĻ½è¡£
    0.09
    ãģĦãģŁ
    0.09
    è¯Ľ
    0.09
    list
    0.09
    []{"
    0.09
    à¸ĩ
    0.09
    Activations Density 0.003%

    No Known Activations