© 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. 10107
    Prev
    Next
    INDEX
    Explanations

    lead roles, supporting roles, title role

    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
    åıijè¾¾
    -0.09
    åIJİèĢħ
    -0.09
    ç«ŀ
    -0.09
    ervo
    -0.08
    ä¼Ĭå§ĭ
    -0.08
    oa
    -0.08
    ä½Ļ家
    -0.08
    AO
    -0.08
    åĻ¤
    -0.08
     sic
    -0.08
    POSITIVE LOGITS
    è§Ĵèī²
    0.34
     roles
    0.30
    çļĦè§Ĵèī²
    0.29
     role
    0.29
     ÑĢоли
    0.26
     ÑĢолÑĮ
    0.25
     Role
    0.25
    role
    0.23
     Roles
    0.23
    roles
    0.23
    Activations Density 0.130%

    No Known Activations