© 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. 2543
    Prev
    Next
    INDEX
    Explanations

    second occurrence

    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¸ĢèĪ¬çļĦ
    -0.11
     tslib
    -0.10
    LOUR
    -0.09
    зв
    -0.09
     khó
    -0.09
    éĥ½ä¸įæķ¢
    -0.09
    rops
    -0.08
    Fact
    -0.08
    æķĿ
    -0.08
     Bernardino
    -0.08
    POSITIVE LOGITS
    çĶµåĻ¨
    0.10
    åĻİ
    0.09
    åį·
    0.08
    è§ģè¯ģ
    0.08
    è¿Ļ家
    0.08
    è¿Ļ家åħ¬åı¸
    0.08
    èĩªè´¸
    0.08
    ä¸Ģ大
    0.08
    ifest
    0.08
    reff
    0.08
    Activations Density 0.057%

    No Known Activations