© 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. 33314
    Prev
    Next
    INDEX
    Explanations

    intermediate risk or level

    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
    ÑĬ
    -0.09
    ners
    -0.09
    SSID
    -0.08
    itian
    -0.08
     Hardware
    -0.08
    ÙĪØ©
    -0.08
     hardware
    -0.08
    è·Łè¸ª
    -0.08
    åĽ±
    -0.07
    nar
    -0.07
    POSITIVE LOGITS
    è¿ĺæľīä»Ģä¹Ī
    0.10
    xef
    0.10
    å°ijæķ°
    0.10
    angler
    0.09
    _mC
    0.09
    为主
    0.09
    å¿ĥ缮ä¸Ń
    0.08
    Äĭ
    0.08
    çļĦ好
    0.08
    æ¯ıä¸Ģ个人
    0.08
    Activations Density 0.004%

    No Known Activations