© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21060
    Prev
    Next
    INDEX
    Explanations

    which accompanies

    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
    éļį
    -0.10
    ctors
    -0.10
    oyer
    -0.09
    ying
    -0.09
    ãģĤãĤĭ
    -0.09
    å§Ķåī¯ä¹¦è®°
    -0.09
    istically
    -0.09
     вÑģÑıк
    -0.09
    sequ
    -0.09
    ä¸Ńæľī
    -0.09
    POSITIVE LOGITS
    soever
    0.14
    /how
    0.12
    opi
    0.10
    ãģ¹ãģį
    0.09
    /if
    0.09
    edor
    0.08
    HING
    0.08
    otta
    0.08
    EVER
    0.08
    mel
    0.08
    Activations Density 0.191%

    No Known Activations