© 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. 4820
    Prev
    Next
    INDEX
    Explanations

    image primitives

    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
    è¹´
    -0.10
    污泥
    -0.09
    éĹŃ
    -0.09
    RTL
    -0.09
    boro
    -0.09
    hua
    -0.09
    &T
    -0.09
    TRL
    -0.09
    å°ģéĶģ
    -0.09
    ULT
    -0.09
    POSITIVE LOGITS
    æĩĭ
    0.09
     primitives
    0.08
     signed
    0.08
    quil
    0.08
    åĴĮå¹³
    0.08
     Berk
    0.08
    为ä¹ĭ
    0.08
     elabor
    0.07
    éĻ
    0.07
    aps
    0.07
    Activations Density 0.153%

    No Known Activations