© 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. 57737
    Prev
    Next
    INDEX
    Explanations

    Like an

    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¸ĢåįĬ
    -0.11
    _NOTICE
    -0.10
    ิà¸ķ
    -0.10
     NOTICE
    -0.09
    çĥģ
    -0.09
    _STAGE
    -0.09
    çīĸ
    -0.09
    _tunnel
    -0.09
    éĻ¤æѤ
    -0.09
    顺便
    -0.09
    POSITIVE LOGITS
    èĢĮæĿ¥
    0.09
     monday
    0.09
    æľ«
    0.08
    ä¸ĩ人
    0.08
     supervised
    0.08
     but
    0.08
    place
    0.08
    intr
    0.08
    atar
    0.08
    boo
    0.08
    Activations Density 0.043%

    No Known Activations