© 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. 50478
    Prev
    Next
    INDEX
    Explanations

    streak

    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
    æĺ¥ç§ĭ
    -0.09
    ague
    -0.09
    tant
    -0.09
    ä¸įä¸ĭ
    -0.08
    odel
    -0.08
    åĪ¿
    -0.08
    cation
    -0.08
    羣æŃ£åģļåĪ°
    -0.08
    ities
    -0.08
    æĺ§
    -0.08
    POSITIVE LOGITS
    ingly
    0.17
    ../../../
    0.10
    ä¸ĭçıŃ
    0.10
    light
    0.09
    ple
    0.09
    ened
    0.09
    utz
    0.09
    hari
    0.09
    ãĥĪãĤ¤
    0.09
    alytics
    0.09
    Activations Density 0.003%

    No Known Activations