© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9461
    Prev
    Next
    INDEX
    Explanations

    loose .dtd or loose words

    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
    aceous
    -0.11
    à¹Į
    -0.11
    UnderTest
    -0.09
    soever
    -0.09
    itious
    -0.09
    ãģĦãģ¾ãģĻ
    -0.08
     dolayı
    -0.08
    ization
    -0.08
    ically
    -0.08
    dsa
    -0.08
    POSITIVE LOGITS
    ening
    0.13
    NING
    0.11
    NESS
    0.11
    est
    0.11
     minded
    0.11
    vé
    0.11
    ened
    0.10
    eners
    0.10
    à¹Ĩ
    0.10
    wares
    0.10
    Activations Density 0.004%

    No Known Activations