© 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. 22646
    Prev
    Next
    INDEX
    Explanations

    picture:

    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
    å°¬
    -0.09
    ith
    -0.09
    åĸĩ
    -0.08
    çļĭ
    -0.08
    æĮ¤
    -0.08
    ä¸ĵ注
    -0.08
    åĬ©æīĭ
    -0.08
    Gam
    -0.08
    翻身
    -0.08
     fingerprint
    -0.08
    POSITIVE LOGITS
    axe
    0.10
    满åĪĨ
    0.10
    agged
    0.10
    çĸ®
    0.09
    .Flags
    0.09
    ured
    0.09
    INGTON
    0.09
    é¨İ
    0.09
    éĢĥçĶŁ
    0.09
    æį¨ãģ¦
    0.09
    Activations Density 0.049%

    No Known Activations