© 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. 34156
    Prev
    Next
    INDEX
    Explanations

    hidden meaning or knowledge

    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
    estruction
    -0.09
    .media
    -0.09
    txn
    -0.08
    人为
    -0.08
    EmptyEntries
    -0.08
    æĵį
    -0.08
    owitz
    -0.08
    2
    -0.08
    9
    -0.08
    4
    -0.08
    POSITIVE LOGITS
    éļIJèĹı
    0.18
    èĹıçĿĢ
    0.17
    (hidden
    0.16
    èĹı
    0.15
    éļIJ
    0.15
    åĮ¿
    0.14
     hidden
    0.14
    éļ±
    0.14
    éļIJèĶ½
    0.14
     hiding
    0.13
    Activations Density 0.060%

    No Known Activations