© 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. 18448
    Prev
    Next
    INDEX
    Explanations

    will followed by a verb

    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
    åı¯æĥ³
    -0.12
    æĺ¯éĿŀ
    -0.10
    èĥ½å¾ĹåĪ°
    -0.09
    .isSuccessful
    -0.08
     pitches
    -0.08
    ï¸
    -0.08
    åĪ°æŶåĢĻ
    -0.08
     mon
    -0.08
     INLINE
    -0.08
    æģIJæĢķ
    -0.08
    POSITIVE LOGITS
    å¾Ģå¾Ģ
    0.12
    å¾Ģå¾Ģæĺ¯
    0.10
    sometimes
    0.10
    å¾Ģå¾Ģä¼ļ
    0.09
    æĬ¤
    0.09
    ary
    0.09
     typically
    0.09
    ahr
    0.09
    ieri
    0.09
    rush
    0.09
    Activations Density 0.085%

    No Known Activations