© 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. 36389
    Prev
    Next
    INDEX
    Explanations

    assassin, assumption, assessment

    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
    uality
    -0.10
    ivity
    -0.10
    idad
    -0.10
    沧æ¡ij
    -0.09
    IZER
    -0.09
    anness
    -0.09
    ä¹Łä¸įæĥ³
    -0.09
    زÙĨ
    -0.09
    ivate
    -0.09
    зов
    -0.08
    POSITIVE LOGITS
    essment
    0.12
    ðŁı»
    0.11
    orted
    0.11
    ignment
    0.11
    ments
    0.10
    ðŁı¼
    0.10
    umption
    0.10
    cess
    0.10
    æĸ
    0.10
    ses
    0.09
    Activations Density 0.041%

    No Known Activations