© 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. 32301
    Prev
    Next
    INDEX
    Explanations

    Marshall

    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
    ally
    -0.11
    ual
    -0.10
    交æ±ĩ
    -0.10
    inally
    -0.09
    lessly
    -0.09
    å§Ķä¼ļ
    -0.09
    ening
    -0.09
    finity
    -0.08
    ular
    -0.08
    楮
    -0.08
    POSITIVE LOGITS
    roe
    0.11
    atten
    0.11
    ses
    0.10
    lest
    0.10
    аÑħ
    0.09
    vÃŃ
    0.09
    们çļĦ
    0.09
    arters
    0.09
    à¹Ħà¸Ĺย
    0.09
    engers
    0.09
    Activations Density 0.006%

    No Known Activations