© 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. 1715
    Prev
    Next
    INDEX
    Explanations

    diagnosis and management

    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
    tion
    -0.10
    ष
    -0.10
    tips
    -0.09
    èģĶ
    -0.08
    oving
    -0.08
    æ¸ħæ°´
    -0.08
    çģ°èī²
    -0.08
    hint
    -0.08
    t
    -0.08
    Ñģп
    -0.08
    POSITIVE LOGITS
    ses
    0.12
    éĮĦ
    0.08
    (es
    0.08
    åĽ½åº¦
    0.08
    å¾ĹåĪ°æľīæķĪ
    0.08
    åŀ¸
    0.08
    æĥħåĨµè¿Ľè¡Į
    0.08
    imple
    0.08
    åį±å®³
    0.08
    éĦ£
    0.08
    Activations Density 0.018%

    No Known Activations