© 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. 38619
    Prev
    Next
    INDEX
    Explanations

    episode air date

    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
    -0.14
    point
    -0.10
    ĺ
    -0.10
    est
    -0.10
    Utf
    -0.09
    оÑģÑĮ
    -0.09
    paces
    -0.09
    озв
    -0.09
    room
    -0.09
    ish
    -0.09
    POSITIVE LOGITS
    opro
    0.11
    plen
    0.10
    entre
    0.10
    éĻIJãĤĬ
    0.09
    owell
    0.09
    afen
    0.09
    odata
    0.09
    aths
    0.09
     اÙĦÙĥرÙĬÙħ
    0.09
    matic
    0.09
    Activations Density 0.017%

    No Known Activations