© 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. 52669
    Prev
    Next
    INDEX
    Explanations

    atypical

    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
    ä¼ī
    -0.10
    åĪ¶å®ļäºĨ
    -0.09
    ÐIJÐŀ
    -0.09
    ducted
    -0.09
    .ur
    -0.09
    stag
    -0.09
    IRONMENT
    -0.08
    å®Ľå¦Ĥ
    -0.08
    ared
    -0.08
    åĢĮ
    -0.08
    POSITIVE LOGITS
    rees
    0.11
    kinson
    0.10
    ac
    0.09
    ãĤ¥
    0.09
    èĦ¸
    0.08
    åĬªåĬĽ
    0.08
    on
    0.08
    ar
    0.08
    éĩįçĤ¹å·¥ä½ľ
    0.08
    com
    0.08
    Activations Density 0.080%

    No Known Activations