© 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. 28040
    Prev
    Next
    INDEX
    Explanations

    diary entry

    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¸įäºĨ
    -0.10
    éľĦ
    -0.10
    çĿ¥
    -0.10
    éĦĻ
    -0.09
    ible
    -0.09
    åĻ¨æĿIJ
    -0.08
    å¿ĥæĻº
    -0.08
    ve
    -0.08
    åħ·ä½ĵæĥħåĨµ
    -0.08
    erville
    -0.08
    POSITIVE LOGITS
    èĽ¾
    0.10
    _ENTRIES
    0.10
    -like
    0.09
    å¼ıçļĦ
    0.09
    istic
    0.09
    ElementType
    0.09
    èĤ¡
    0.09
    entries
    0.09
    åıįçľģ
    0.09
    /story
    0.08
    Activations Density 0.009%

    No Known Activations