© 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. 52675
    Prev
    Next
    INDEX
    Explanations

    relative to

    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
    aga
    -0.10
    rial
    -0.10
    iers
    -0.09
    line
    -0.08
    ari
    -0.08
    fold
    -0.08
    trie
    -0.08
    aque
    -0.08
    rella
    -0.08
    dream
    -0.08
    POSITIVE LOGITS
    èĢĮè¨Ģ
    0.12
    ìłģìľ¼ë¡ľ
    0.11
    ities
    0.11
    ResourceManager
    0.11
    æĪIJè´¥
    0.11
    湿度
    0.10
    å§Ĭ妹
    0.10
    ä¾Ĩ說
    0.10
    ately
    0.10
    ivent
    0.10
    Activations Density 0.049%

    No Known Activations