© 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. 1577
    Prev
    Next
    INDEX
    Explanations

    entry fee

    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
    ç»ĻåĪ«äºº
    -0.09
    è§
    -0.08
    ANGLE
    -0.08
    ä¸ĵéŨ
    -0.08
    OPS
    -0.08
    èŀį
    -0.07
    ç²Ĺ
    -0.07
    缩水
    -0.07
    Ø·ÙĪ
    -0.07
    ddy
    -0.07
    POSITIVE LOGITS
    loor
    0.11
    æ§Ľ
    0.10
    åŀ¸
    0.10
    addock
    0.10
    aminer
    0.09
    å±ĬæŶ
    0.09
    误
    0.09
    交éĢļè¿IJè¾ĵ
    0.09
    å½ĵåľº
    0.09
    éĩįå·¥
    0.09
    Activations Density 0.029%

    No Known Activations