© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22087
    Prev
    Next
    INDEX
    Explanations

    essential

    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
    zilla
    -0.09
    åij¤
    -0.09
    rell
    -0.09
    佬
    -0.08
     sake
    -0.08
    teenth
    -0.08
    walk
    -0.08
    åķ¸
    -0.08
    sticks
    -0.08
    çīĸ
    -0.08
    POSITIVE LOGITS
    ities
    0.12
    ist
    0.12
    ipse
    0.11
     oils
    0.11
    ìłģìĿ¸
    0.11
    ists
    0.11
    iveness
    0.10
    ized
    0.10
    ment
    0.10
    iy
    0.10
    Activations Density 0.022%

    No Known Activations