© 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. 52680
    Prev
    Next
    INDEX
    Explanations

    identifying or characterization

    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
    ishly
    -0.11
    iness
    -0.10
    ter
    -0.10
    ìį¨
    -0.09
    adero
    -0.09
    ÙĤØ·
    -0.09
    pedia
    -0.09
    ipa
    -0.09
    ç¯ģ
    -0.09
    atten
    -0.08
    POSITIVE LOGITS
    符
    0.12
    ENTITY
    0.10
    entifier
    0.10
    å¢ħ
    0.10
     ÑģебÑı
    0.09
    èĦĪ
    0.09
    zelf
    0.09
    çijķ
    0.09
    ATAR
    0.09
    ç½ijå°ıç¼ĸ
    0.09
    Activations Density 0.048%

    No Known Activations