© 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. 6693
    Prev
    Next
    INDEX
    Explanations

    Copyright lines and authors

    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
    ä¸įçķĻ
    -0.11
    åĢľ
    -0.10
    æ·ĭ
    -0.09
    okes
    -0.09
    迤
    -0.09
    éģ´
    -0.09
    çľī
    -0.09
    éªļ
    -0.09
    æijĬ
    -0.09
    çķĻåľ¨
    -0.09
    POSITIVE LOGITS
    iture
    0.08
    icable
    0.08
     Báo
    0.08
    éĤ´
    0.08
    ç¿Ģ
    0.08
    Audit
    0.08
     PM
    0.08
    erset
    0.08
    audi
    0.08
    æĭľå¸Ī
    0.08
    Activations Density 0.346%

    No Known Activations